WO2023060592A1 - 定位方法及装置 - Google Patents

定位方法及装置 Download PDF

Info

Publication number
WO2023060592A1
WO2023060592A1 PCT/CN2021/124206 CN2021124206W WO2023060592A1 WO 2023060592 A1 WO2023060592 A1 WO 2023060592A1 CN 2021124206 W CN2021124206 W CN 2021124206W WO 2023060592 A1 WO2023060592 A1 WO 2023060592A1
Authority
WO
WIPO (PCT)
Prior art keywords
grid
information
attribute information
environment
local
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/124206
Other languages
English (en)
French (fr)
Inventor
赵小文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP21960322.2A priority Critical patent/EP4404018A4/en
Priority to CN202180103257.8A priority patent/CN118119907A/zh
Priority to PCT/CN2021/124206 priority patent/WO2023060592A1/zh
Publication of WO2023060592A1 publication Critical patent/WO2023060592A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/005—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 with correlation of navigation data from several sources, e.g. map or contour matching
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
    • G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
    • G01C21/1652—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments with ranging devices, e.g. LIDAR or RADAR
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/20—Control system inputs
    • G05D1/24—Arrangements for determining position or orientation
    • G05D1/242—Means based on the reflection of waves generated by the vehicle
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/20—Control system inputs
    • G05D1/24—Arrangements for determining position or orientation
    • G05D1/246—Arrangements for determining position or orientation using environment maps, e.g. simultaneous localisation and mapping [SLAM]
    • G05D1/2464—Arrangements for determining position or orientation using environment maps, e.g. simultaneous localisation and mapping [SLAM] using an occupancy grid
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/88—Lidar systems specially adapted for specific applications
    • G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
    • G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2107/00—Specific environments of the controlled vehicles
    • G05D2107/10—Outdoor regulated spaces
    • G05D2107/13—Spaces reserved for vehicle traffic, e.g. roads, regulated airspace or regulated waters
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2109/00—Types of controlled vehicles
    • G05D2109/10—Land vehicles
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D2111/00—Details of signals used for control of position, course, altitude or attitude of land, water, air or space vehicles
    • G05D2111/10—Optical signals
    • G05D2111/17—Coherent light, e.g. laser signals

Definitions

  • the present application relates to the technical field of positioning, and in particular to a positioning method and device.
  • Intelligent driving includes assisted driving and automatic driving.
  • Their key technologies include positioning, perception, planning and decision-making, control and so on.
  • high-precision positioning can provide a reliable basis for planning decision-making or control. Therefore, high-precision positioning plays an indispensable role in intelligent driving.
  • GPS global positioning system
  • Embodiments of the present application provide a positioning method and device, which can improve positioning accuracy and stability.
  • a positioning method includes: acquiring a grid map corresponding to the first environment, acquiring a local grid corresponding to the second environment, and acquiring when the local grid is located at a different position in the grid map , the first matching probability between the local grid and the grid in the grid map, and determine the position of the device to be located according to the first matching probability.
  • the above-mentioned grid map includes a plurality of first grids, and any first grid includes a plurality of points, location information of the plurality of points, and attribute information of the first grid.
  • the attribute information of the first grid includes semantic information of the first grid and height information of the first grid, and the semantic information of the first grid is used to indicate that the first grid represents each of various objects.
  • the height information of the first grid is used to indicate the height of the points included in the first grid;
  • the second environment is included in the first environment, the local grid includes at least one second grid , any second grid includes a plurality of points and the attribute information of the second grid, the attribute information of the second grid includes the semantic information of the second grid and the height information of the second grid, the first
  • the semantic information of the second grid is used to indicate the probability that the second grid represents each object in the plurality of objects, and the height information of the second grid is used to indicate the height of the points included in the second grid.
  • the device to be located is located in the second environment.
  • the positioning device can obtain the grid map corresponding to the first environment, obtain the local grid corresponding to the second environment information, and combine various attribute information of the first grid with the second grid A variety of attribute information of the positioning device is used to locate the positioning device, which improves the accuracy and stability of positioning.
  • the attribute information of the first grid further includes reflection intensity information of the first grid, and the reflection intensity information of the first grid is used to indicate the Reflection intensity
  • the attribute information of the second grid further includes reflection intensity information of the second grid, and the reflection intensity information of the second grid is used to indicate the reflection intensity of the points included in the second grid.
  • the positioning device can also combine the reflection intensity information of the first grid and the reflection intensity information of the second grid to perform positioning, further improving the accuracy and stability of positioning.
  • the method further includes: predicting the position corresponding to the position of the device to be positioned at the second moment according to the pose information of the device to be positioned at the first moment and the motion parameters of the device to be positioned. an area. Based on the above method, the positioning device can predict the position of the device to be positioned at the second moment, so as to position the device to be positioned according to the predicted position.
  • the acquiring the first matching probability includes: when acquiring the grid where the local grid is located in the first area, the first matching probability between the local grid and the grid in the grid map match probability. Based on the above method, the positioning device does not need to match the grids in the local grid with all the grids in the grid map, but matches the grids in the local grid with the grids in the first region, Reduced computational overhead.
  • the first matching probability between the local grid and the grid in the grid map includes: according to the local grid The attribute information of the grids in the grid and the attribute information of the grids in the first area are used to obtain the first matching probability corresponding to each type of attribute information of the first grids in the first area.
  • the positioning device can determine the first matching probability corresponding to each type of attribute information of the first grid in the first area based on the attribute information of the first grid and the attribute information of the second grid, and then can according to the first The matching probability is used to locate the positioning device, so that the positioning device does not rely only on a certain attribute for positioning during positioning, which improves the accuracy and stability of positioning.
  • the positioning device can match each second grid in the local grid with each first grid in the first area to determine the first attribute information of the first grid in the first area The corresponding first matching probability makes the positioning accuracy of the device to be positioned higher.
  • determining the position of the device to be positioned according to the first matching probability includes: determining each attribute according to the first matching probability corresponding to each attribute information of the first grid in the first area The location information corresponding to the information; according to the location information corresponding to each attribute information, the location of the device to be positioned is determined. Based on the above method, the positioning device can obtain the location information corresponding to each type of attribute information according to the first matching probability corresponding to each type of attribute information of the first grid in the first area, and obtain the location of the device to be positioned according to the location information , so as to locate the device to be positioned.
  • determining the position of the device to be positioned according to the first matching probability further includes: determining the location information corresponding to at least one type of attribute information conforming to preset rules in the location information corresponding to each type of attribute information.
  • Position information according to the position information corresponding to each type of attribute information, determining the position of the device to be positioned includes: if the quantity of the at least one type of attribute information is greater than 1, weighting the position information corresponding to the at least one type of attribute information and, obtaining the location of the device to be located; or, if the quantity of the at least one type of attribute information is equal to 1, determining the location information corresponding to the at least one type of attribute information as the location of the device to be located.
  • the positioning device determines the location information corresponding to each attribute information, it can first screen the location information, eliminate inaccurate location information, and then determine the location of the device to be positioned according to the filtered location information, so as to This enables the positioning device to more accurately locate the positioning device to be positioned.
  • the method further includes: acquiring a second matching probability, where the second matching probability is when the local grid is located at the position of the device to be positioned and the local grid is rotated by different angles , the matching probability of the local grid and the grid in the grid map; determine the orientation angle of the device to be positioned according to the second matching probability.
  • the positioning device can determine the orientation angle of the device to be positioned, that is, the moving direction of the device to be positioned.
  • the method further includes: predicting the orientation angle of the device to be positioned at the second moment according to the orientation angle of the device to be positioned at the third moment and the motion parameters of the device to be positioned. Based on the above method, the positioning device can predict the orientation angle of the device to be positioned at the second moment, so as to determine the orientation angle of the device to be positioned according to the predicted orientation angle.
  • obtaining the second matching probability includes: obtaining the second matching probabilities of multiple rotation angles corresponding to the local grid when the local grid is located at the position of the device to be positioned, Among the plurality of rotation angles, the included angle between each rotation angle and the predicted orientation angle of the device to be positioned at the second moment is less than or equal to the first angle.
  • the positioning device does not need to rotate the local grid once to obtain multiple rotation angles, but rotates the local grid around the orientation angle at the second moment to obtain multiple rotation angles. It can not only reduce the calculation cost of the positioning device, but also improve the accuracy of the determined orientation angle of the device to be positioned.
  • obtaining the second matching probability of the first rotation angle when the local grid is located at the position of the device to be positioned includes: The attribute information of the grid and the attribute information of the grid in the grid map, and obtain the second matching probability corresponding to each attribute information of the first rotation angle, and the first rotation angle is any of the plurality of rotation angles a rotation angle.
  • the positioning device can determine the second matching probability corresponding to each attribute information of the first rotation angle based on the attribute information of the first grid and the attribute information of the second grid, and then determine the undetermined probability according to the second matching probability.
  • the orientation angle of the positioning device makes the positioning device not only rely on a certain attribute when determining the orientation angle of the device to be positioned, and improves the accuracy of the orientation angle determined by the positioning device.
  • the first attribute of the first rotation angle is acquired according to the attribute information of the grid in the local grid corresponding to the first rotation angle and the attribute information of the grid in the grid map
  • the second matching probability corresponding to the information includes: obtaining the position of each second grid in the partial grid when the local grid is located at the position of the device to be positioned and the partial grid is rotated by a first rotation angle
  • the first attribute information, the second deviation between the first attribute information of the first grid corresponding to the second grid, the first attribute information of the second grid is the attribute information of the second grid
  • the first attribute information of the first grid is the same as the first attribute information of the second grid; according to the second deviation, the first attribute information corresponding to the first rotation angle is obtained.
  • the positioning device can match each second grid in the local grid with the first grid corresponding to the second grid at the first rotation angle, so as to determine the first attribute information of the first rotation angle
  • the corresponding second matching probability makes the orientation angle determined by the positioning device more accurate.
  • determining the orientation angle of the device to be positioned according to the second matching probability includes: determining the orientation corresponding to each attribute information according to the second matching probability corresponding to each attribute information of each rotation angle
  • Angle information the orientation angle information corresponding to each attribute information is weighted and summed to obtain the orientation angle of the device to be positioned.
  • the positioning device can obtain the orientation angle information corresponding to each attribute information according to the second matching probability corresponding to each attribute information of each rotation angle, and obtain the orientation angle of the device to be positioned according to the orientation angle information, Thus, the positioning device can determine the moving direction of the device to be positioned.
  • acquiring the grid map corresponding to the first environment includes: acquiring point cloud information corresponding to the first environment, where the point cloud information corresponding to the first environment is used to indicate the first environment The location information of the points in the corresponding point cloud; according to the point cloud information corresponding to the first environment, the grid map corresponding to the first environment is obtained. Based on the above method, the positioning device can acquire the grid map corresponding to the first environment.
  • the information of the point cloud corresponding to the first environment is further used to indicate the reflection intensity of points in the point cloud corresponding to the first environment.
  • the positioning device may acquire reflection intensity information of the first grid according to point cloud information corresponding to the first environment.
  • obtaining the local grid corresponding to the second environment includes: obtaining point cloud information corresponding to the second environment, where the point cloud information corresponding to the second environment is used to indicate the second environment The location information of the points in the corresponding point cloud; according to the point cloud information corresponding to the second environment, the local grid corresponding to the second environment is obtained. Based on the above method, the positioning device can acquire the local grid corresponding to the second environment.
  • the information of the point cloud corresponding to the second environment is also used to indicate the reflection intensity of points in the point cloud corresponding to the second environment.
  • the positioning device may acquire reflection intensity information of the second grid according to point cloud information corresponding to the second environment.
  • a positioning device for implementing the above method.
  • the positioning device may be the positioning device in the above first aspect, or a device including the above positioning device.
  • the positioning device includes corresponding modules, units, or means (means) for realizing the above method, and the modules, units, or means can be implemented by hardware, software, or by executing corresponding software on hardware.
  • the hardware or software includes one or more modules or units corresponding to the above functions.
  • a positioning device including: a processor; the processor is configured to be coupled with a memory, and after reading an instruction in the memory, execute the method according to the above first aspect according to the instruction.
  • the positioning device may be the positioning device in the above first aspect, or a device including the above positioning device.
  • the positioning device further includes a memory, and the memory is configured to store necessary program instructions and data.
  • the positioning device is a chip or a chip system.
  • the positioning device when it is a system-on-a-chip, it may consist of chips, or may include chips and other discrete devices.
  • a positioning device including: a processor and an interface circuit; the interface circuit is used to receive a computer program or instruction and transmit it to the processor; the processor is used to execute the computer program or instruction, so that the positioning The device executes the method described in the first aspect above.
  • the positioning device is a chip or a chip system.
  • the positioning device when it is a system-on-a-chip, it may consist of chips, or may include chips and other discrete devices.
  • a computer-readable storage medium is provided, and instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer can execute the method described in the above-mentioned first aspect.
  • a computer program product containing instructions is provided, and when it is run on a computer, the computer can execute the method described in the first aspect above.
  • the technical effects brought about by any one of the possible implementations from the second aspect to the sixth aspect can refer to the technical effects brought about by the different possible implementations in the above-mentioned first aspect or any one of the aspects, and no longer repeat.
  • FIG. 1A is a first schematic diagram of a positioning system architecture provided by an embodiment of the present application.
  • FIG. 1B is a second schematic diagram of the positioning system architecture provided by the embodiment of the present application.
  • FIG. 2 is a schematic diagram of the hardware structure of the positioning device provided by the embodiment of the present application.
  • FIG. 3 is a first schematic flow diagram of a positioning method provided in an embodiment of the present application.
  • FIG. 4 is a first schematic diagram of a grid map and a local grid provided by the embodiment of the present application.
  • FIG. 5 is a schematic flow diagram II of the positioning method provided by the embodiment of the present application.
  • FIG. 6 is a schematic diagram of the angular velocity of the device to be positioned, the linear velocity of the device to be positioned and the rotation angle of the wheel of the device to be positioned according to the embodiment of the present application;
  • FIG. 7 is a second schematic diagram of a grid map and a partial grid provided by the embodiment of the present application.
  • FIG. 8 is a schematic flow diagram III of the positioning method provided by the embodiment of the present application.
  • FIG. 9 is a third schematic diagram of a grid map and a local grid provided by the embodiment of the present application.
  • FIG. 10 is a fourth schematic flow diagram of the positioning method provided by the embodiment of the present application.
  • FIG. 11 is a schematic flow diagram five of the positioning method provided by the embodiment of the present application.
  • Fig. 12 is a schematic diagram of the position of the device to be positioned provided by the embodiment of the present application.
  • FIG. 13 is a schematic structural diagram of a positioning device provided by an embodiment of the present application.
  • SLAM can specify positioning and map construction, or concurrent mapping and positioning.
  • a map acquisition device such as a robot
  • the map acquisition device can gradually draw the map while moving according to the surrounding environment information and the position information of the map acquisition device at each time. Environment map.
  • the ICP algorithm can determine the point set Qi closest to the point set Pi in the target point cloud set in the source point cloud set according to certain constraints, and calculate the rotation matrix R and translation matrix T, so that the error function is minimized; Use R and T to perform rotation and translation transformation to obtain a new point P' i ; calculate the average distance d between P' i and Qi, if d is less than a certain threshold, or the number of iterations to calculate the rotation matrix and translation matrix is greater than the maximum number of iterations, Then stop the calculation, otherwise re-determine the point set Q' i closest to the point set Pi in the target point cloud set in the source point cloud set, and continue to iteratively calculate the rotation matrix and translation matrix until the obtained average distance is less than a certain Threshold, or the number of iterations to calculate the rotation matrix and translation matrix is greater than the maximum number of iterations.
  • LiDAR odometry and mapping is an algorithm for implementing SLAM using LiDAR.
  • the position of the robot relative to the initial moment is calculated by matching the lidar point cloud, which is called the lidar odometer. According to each position, the point cloud is combined to form a map, which is called mapping.
  • PCL is a large-scale cross-platform open source C++ programming library established on the basis of absorbing previous point cloud related research. It implements a large number of point cloud-related general algorithms and efficient data structures, involving point cloud acquisition, filtering, segmentation, registration, retrieval, feature extraction, recognition, tracking, surface reconstruction, visualization, etc. It supports multiple operating system platforms and can run on Windows, Linux, Android, Mac OS X, and some embedded real-time systems.
  • PointNet is a new type of deep learning model for processing point cloud data. It can be used for various cognitive tasks on point cloud data, such as classification, semantic segmentation and object recognition.
  • the following method can be used for positioning to improve the accuracy and stability of positioning.
  • the ICP algorithm can be used to combine the point cloud (which can be equivalent to the above-mentioned Pi) obtained by the lidar in real time with the point cloud in the pre-established point cloud map (which can be equivalent to the point set Pi in the above-mentioned and target point cloud collection) The closest point set) to match to determine the real-time position.
  • Method 2 Positioning method based on feature information and grid map
  • the local grid can be obtained according to the point cloud information (such as the position information of the point cloud, the feature information of the point cloud, etc.) obtained by the lidar in real time, and the feature information of each grid in the local grid
  • the feature information of the point cloud in the raster map is matched to complete the positioning.
  • the feature information of the point cloud may include height information.
  • the pre-stored grid map is obtained from the point cloud information obtained by the lidar in the environment corresponding to the grid map.
  • Method 3 Positioning method based on category information and grid map
  • the local grid can be obtained according to the position information of the point cloud obtained by the lidar in real time, and the image information corresponding to the above point cloud obtained by the visual sensor, and the object in the image information can be identified to obtain the type information corresponding to the object. Save the type information in the local grid; then match the type information of each grid in the local grid with the type information of each grid in the pre-stored grid map, so as to complete the positioning.
  • the pre-stored grid map is obtained according to the point cloud information acquired by the laser radar in the environment corresponding to the grid map, and the image information acquired by the visual sensor in the environment corresponding to the grid map.
  • Method 4 Positioning method based on attribute information and grid map
  • a grid map corresponding to the first environment may be obtained, and a local grid corresponding to the second environment information may be obtained.
  • the grid map includes a plurality of first grids, the first grid includes a plurality of points, the position information of each point in the plurality of points, and the attribute information of the first grid, and the attribute information of the first grid includes the first The semantic information of the first grid and the height information of the first grid.
  • the second environment includes that in the first environment, the local grid includes at least one second grid, and the second grid includes a plurality of points and attribute information of the second grid.
  • the attribute information of the second grid includes semantic information of the second grid and height information of the second grid.
  • the first matching probability between the local grid and the grid in the grid map may be obtained, and the position of the device to be positioned is determined according to the first matching probability.
  • Method 4 will be described in the following embodiment shown in FIG. 3 , and will not be repeated here.
  • the above method 1 is relatively simple to implement, method 1 only considers the minimum distance between points, and the algorithm tends to converge to a wrong position, that is, the positioning is inaccurate.
  • the above method 2 combines the feature information of the point cloud for positioning, which improves the positioning accuracy compared with method 1. However, if the characteristic information changes, the positioning of method 2 will be inaccurate and the stability will be poor.
  • the above method 3 combines the type information of the object for positioning, and compared with method 1, the positioning accuracy is improved. However, method 3 generally relies on the accuracy of recognizing objects in image information or point cloud information, and if the accuracy of recognizing objects is low, the positioning will be inaccurate.
  • the above method 4 can be combined with the attribute information of the grid, such as the height information of the grid and the semantic information of the grid for positioning. Compared with methods 1-3, the positioning accuracy and stability of method 4 are higher.
  • the embodiments of the present application provide a positioning method and device, which can perform positioning within a certain area.
  • the positioning method and device can be used for positioning in environments such as residential quarters, industrial parks, and parks.
  • the positioning accuracy and stability can be improved.
  • the applicable system architecture of this embodiment of the present application includes a positioning device.
  • the system architecture also includes a sensing device.
  • the sensing device can communicate with the positioning device in a wired or wireless manner.
  • the system architecture may be as shown in FIG. 1A.
  • the system architecture 10 shown in FIG. 1A includes a positioning device 101 .
  • the system architecture 10 further includes a sensing device 102 .
  • the above-mentioned positioning device can be used to obtain the grid map corresponding to the first environment, obtain the local grid corresponding to the second environment, and obtain the local grid and grid map when the local grid is located at different positions in the grid map.
  • the first matching probability of the grid and determine the position of the device to be located according to the first matching probability.
  • the grid map includes a plurality of first grids. Any first grid includes a plurality of points, position information of each point in the plurality of points, and attribute information of the first grid.
  • the attribute information of the first grid includes semantic information of the first grid and height information of the first grid.
  • the second environment is included in the first environment, and the local grids include at least one second grid. Any second grid includes multiple points and attribute information of the second grid.
  • the attribute information of the second grid includes semantic information of the second grid and height information of the second grid.
  • the above positioning apparatus may be a vehicle-mounted device or a computing device.
  • the above-mentioned positioning device is a computer, a computing unit in a computer, an embedded device, an industrial computer of an x86 system, or a wireless terminal in unmanned driving.
  • the above-mentioned positioning device may be a vehicle-mounted module, vehicle-mounted module, vehicle-mounted component, vehicle-mounted chip or vehicle-mounted unit built into the vehicle as one or more components or units.
  • the vehicle uses the built-in vehicle module, vehicle-mounted module, vehicle-mounted component,
  • the on-board chip or on-board unit can implement the method of the present application.
  • the above sensing device may be used to obtain point cloud information corresponding to the first environment, and send the point cloud information to the above positioning device.
  • the point cloud information corresponding to the first environment may be used to indicate position information of points in the point cloud corresponding to the first environment.
  • the information of the point cloud corresponding to the first environment may also be used to indicate the reflection intensity of the points in the point cloud corresponding to the first environment.
  • the sensing device may also be used to acquire point cloud information corresponding to the second environment, and send the point cloud information to the positioning device.
  • the point cloud information corresponding to the second environment is used to indicate the position information of the point cloud corresponding to the second environment.
  • the point cloud information corresponding to the second environment may also be used to indicate the reflection intensity of points in the point cloud corresponding to the second environment.
  • the above sensing device may be a lidar, or other devices capable of acquiring point cloud information corresponding to a certain environment.
  • the applicable system architecture of this embodiment of the present application includes a positioning device.
  • a positioning module can be deployed in the positioning device.
  • a perception module is also deployed in the positioning device.
  • the system architecture may be as shown in FIG. 1B.
  • the system architecture 11 shown in FIG. 1B includes a positioning device 111 .
  • a positioning module 1111 is deployed in the positioning device 111 .
  • a perception module 1112 is further deployed in the positioning device 111 .
  • the above-mentioned positioning module can be used to obtain the grid map corresponding to the first environment, obtain the local grid corresponding to the second environment, and obtain the local grid and grid map when the local grid is located at different positions in the grid map.
  • the first matching probability of the grid and determine the position of the device to be located according to the first matching probability.
  • the above perception module may be used to acquire point cloud information corresponding to the first environment, and send the point cloud information to the above positioning module.
  • the above perception module can also be used to acquire point cloud information corresponding to the second environment, and send the point cloud information to the above positioning module.
  • the above positioning device may be a device having a perception function and a calculation function.
  • the above-mentioned positioning device is a smart car, such as an unmanned car, an automatic takeaway car, an automatic express car, and the like.
  • the positioning module is a module having a calculation function in the positioning device
  • the sensing module is a module having a sensing function in the positioning device.
  • the positioning module can be an on-board computer in a smart car, a computing unit in a car computer, an embedded device in a smart car, an industrial computer in a smart car, or a wireless terminal in a smart car, etc.;
  • the perception module can be the laser radar in the smart car, or other devices in the smart car that can obtain point cloud information corresponding to a certain environment.
  • FIG. 1A and FIG. 1B are only schematic diagrams, and do not constitute a limitation on the application scenarios of the technical solutions provided in this application. Those skilled in the art should understand that in the specific implementation process, the above system architecture may also include other devices, and the number of positioning devices, sensing devices, positioning modules or sensing modules may also be determined according to specific needs, without limitation.
  • the device for example, the positioning device shown in FIG. 1A or FIG. 1B in this embodiment of the present application may be a general-purpose device or a special-purpose device, which is not specifically limited in this embodiment of the present application.
  • the relevant functions of the device (such as the positioning device) in FIG. 1A or FIG. 1B in the embodiment of the present application can be realized by one device, or by multiple devices, or by one or more It is implemented by a functional module, which is not specifically limited in the embodiment of the present application.
  • a functional module which is not specifically limited in the embodiment of the present application.
  • the above-mentioned functions can be network elements in hardware devices, software functions running on dedicated hardware, or a combination of hardware and software, or instantiated virtualization on a platform (for example, a cloud platform) Function.
  • FIG. 2 is a schematic diagram of a hardware structure of a positioning device applicable to an embodiment of the present application.
  • the positioning device 20 includes at least one processor 201, configured to implement the method provided in the embodiment of the present application.
  • the positioning device 20 may also include a communication line 202 , a memory 203 and at least one communication interface 204 .
  • the processor 201 can be a general-purpose central processing unit (central processing unit, CPU), a microprocessor, a specific application integrated circuit (application-specific integrated circuit, ASIC), or one or more for controlling the execution of the program program of this application integrated circuit.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • Communication line 202 may include a path, such as a bus, for transferring information between the aforementioned components.
  • the communication interface 204 is used for communicating with other devices or communication networks.
  • the communication interface 204 can be any device such as a transceiver, such as an Ethernet interface, a wireless access network (radio access network, RAN) interface, a wireless local area network (wireless local area networks, WLAN) interface, a transceiver, a pin , bus, or transceiver circuits, etc.
  • a transceiver such as an Ethernet interface, a wireless access network (radio access network, RAN) interface, a wireless local area network (wireless local area networks, WLAN) interface, a transceiver, a pin , bus, or transceiver circuits, etc.
  • Memory 203 may be read-only memory (read-only memory, ROM) or other types of static storage devices that can store static information and instructions, random access memory (random access memory, RAM) or other types that can store information and instructions It can also be an electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program code in the form of instructions or data structures and can be programmed by a computer Any other medium accessed, but not limited to.
  • the memory may exist independently and be coupled with the processor 201 through the communication line 202 .
  • the memory 203 can also be integrated with the processor 201 .
  • the memory provided by the embodiment of the present application may generally be non-volatile.
  • the memory 203 is used to store computer-executed instructions involved in executing the solutions provided by the embodiments of the present application, and the execution is controlled by the processor 201 .
  • the processor 201 is configured to execute computer-executed instructions stored in the memory 203, so as to implement the method provided in the embodiment of the present application.
  • the processor 201 may also perform processing-related functions in the methods provided in the following embodiments of the application, and the communication interface 204 is responsible for communicating with other devices or communication networks. The example does not specifically limit this.
  • the computer-executed instructions in the embodiments of the present application may also be referred to as application program codes, which is not specifically limited in the embodiments of the present application.
  • the coupling in the embodiments of the present application is an indirect coupling or a communication connection between devices, units or modules, which may be in electrical, mechanical or other forms, and is used for information exchange between devices, units or modules.
  • the processor 201 may include one or more CPUs, such as CPU0 and CPU1 in FIG. 2 .
  • the positioning apparatus 20 may include multiple processors, for example, the processor 201 and the processor 207 in FIG. 2 .
  • Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor.
  • a processor herein may refer to one or more devices, circuits, and/or processing cores for processing data (eg, computer program instructions).
  • composition structure shown in FIG. 2 does not constitute a limitation to the positioning device.
  • some components, or a different arrangement of components are shown in FIG. 2 in addition to the components shown in FIG. 2 in addition to the components shown in FIG. some components, or a different arrangement of components.
  • the positioning method provided by the embodiment of the present application will be described below with reference to the accompanying drawings.
  • the positioning device in the following embodiments may have the components shown in FIG. 2 , which will not be described in detail.
  • the positioning method provided by the embodiment of the present application can be applied in multiple fields, for example: the field of unmanned driving, the field of automatic driving, the field of assisted driving, the field of intelligent driving, the field of networked driving, the field of intelligent networked driving, car sharing, etc.
  • A/B can indicate A or B
  • A/B can indicate A or B
  • and/or can be used to describe There are three kinds of relationships between associated objects, for example, A and/or B, which can mean: A exists alone, A and B exist simultaneously, and B exists independently, where A and B can be singular or plural.
  • expressions like "at least one of A, B, and C" or "at least one of A, B, or C” are generally used to mean any of the following: A alone; B alone; C exists; A and B exist together; A and C exist simultaneously; B and C exist simultaneously; A, B, and C exist simultaneously.
  • the above is an example of the three elements of A, B and C to illustrate the optional items of the item.
  • the meaning of the expression can be obtained according to the aforementioned rules.
  • words such as “first” and “second” may be used to distinguish technical features with the same or similar functions.
  • the words “first” and “second” do not limit the number and execution order, and the words “first” and “second” do not necessarily mean that they must be different.
  • words such as “exemplary” or “for example” are used to represent examples, illustrations or illustrations, and any embodiment or design described as “exemplary” or “for example” should not be interpreted It is more preferred or more advantageous than other embodiments or design solutions.
  • the use of words such as “exemplary” or “for example” is intended to present related concepts in a specific manner for easy understanding.
  • the positioning device may perform some or all of the steps in the embodiment of the present application, and these steps are only examples, and the embodiment of the present application may also perform other steps or variations of various steps.
  • each step may be performed in a different order presented in the embodiment of the present application, and it may not be necessary to perform all the steps in the embodiment of the present application.
  • the positioning method includes S301-S304.
  • the positioning device acquires a grid map corresponding to the first environment.
  • the positioning device may be the positioning device 101 in FIG. 1A , or the positioning device 111 in FIG. 1B .
  • the above-mentioned first environment may be the environment where the device to be positioned is located.
  • the first environment is a residential area, an industrial park, or a park.
  • the device to be positioned is a device that needs to be positioned, and the device to be positioned and the positioning device may be the same or different. If the device to be positioned is the same as the positioning device, it means that the positioning device needs to position itself. If the device to be positioned is different from the positioning device, it means that the positioning device needs to locate a device other than itself.
  • the above grid map is a two-dimensional map.
  • the grid map includes a plurality of first grids.
  • the first grid may be square or rectangular, or some of the plurality of first grids may be square and the other part may be rectangular, which is not limited.
  • the size of the first grid can be fixed, that is, it is pre-stored in the positioning device, or it can be set by the user according to needs.
  • the grid may also have other names, for example, the grid may also be called a grid, an area, etc., which are not limited.
  • the grid map may be as shown in FIG. 4 .
  • the grid map 401 in FIG. 4 includes 63 first grids.
  • the first grid in the grid map 401 is a square.
  • any one of the multiple first grids includes multiple points, position information of each point in the multiple points, and attribute information of the first grid.
  • the location information of the point may be used to indicate the location of the point in the grid map.
  • the position information of a point includes the value x of the abscissa and the value y of the ordinate of the point in the grid map, ie (x, y).
  • the position information of each point included in the first grid may be presented in the first grid by distributing each point in the first grid at different positions of the first grid.
  • the position information of point 1 is (x1, y1)
  • the position information of point 2 is (x2, y2)
  • point 1 and point 2 are at different positions in the first grid.
  • the attribute information of the first grid may include semantic information of the first grid and height information of the first grid.
  • the attribute information of the first grid may also include reflection intensity information of the first grid.
  • Each type of attribute information included in the attribute information of the first grid is introduced below.
  • the semantic information of the first grid may be used to indicate the probability that the first grid represents each of the plurality of objects.
  • the first grid represents the probability of each object in the various objects, which can be understood as the object to which the point in the first grid belongs (or the object corresponding to the point in the first grid) is one of the various objects Probability of each object.
  • the plurality of objects may include at least two of trees, parked vehicles, walls, ground, buildings, lawns, or other objects.
  • the semantic information of the first grid can be used to indicate that the first grid represents the probability of trees, the probability of representing parked vehicles, the probability of representing walls, the probability of representing ground, the probability of representing lawns and the probability of representing other objects.
  • Probability or in other words, the semantic information of the first grid can be used to indicate the probability that the object belonging to the point in the first grid is a tree, a parked vehicle, a wall, a ground, or a lawn The probability of and the probability of other objects.
  • the semantic information of the first grid includes the probability that the first grid represents each of the various objects. Taking a variety of objects including trees, parked vehicles, buildings and other objects as an example, if the probability of the first grid representing trees is 0.95, the probability of representing parked vehicles is 0.02, and the probability of representing buildings is 0.01, representing other The probability of the object is 0.02, and the semantic information of the first grid includes ⁇ 0.95, 0.02, 0.01, 0.02 ⁇ .
  • the semantic information of the first grid includes label information of each of the multiple objects, and a probability that the first grid represents each of the multiple objects. Taking a variety of objects including trees, parked vehicles, buildings and other objects as an example, if the probability of the first grid representing trees is 0.95, the probability of representing parked vehicles is 0.02, and the probability of representing buildings is 0.01, representing other The probability of an object is 0.02, then the semantic information of the first grid includes ⁇ trees: 0.95; parked vehicles: 0.02; buildings: 0.01; other objects: 0.02 ⁇ , or the semantic information of the first grid includes ⁇ trees, 0.95 ⁇ ; ⁇ Parked vehicles: 0.02 ⁇ ; ⁇ Buildings: 0.01 ⁇ ; ⁇ Other objects: 0.02 ⁇ .
  • the semantic information of the first grid may be presented in the first grid through the color attribute parameters of the points in the first grid.
  • the color attribute parameters include at least one of grayscale, saturation, lightness or hue.
  • each object in various objects can correspond to a grayscale value
  • the grayscale value of the point in the first grid is the grayscale value corresponding to the first object
  • the first object is The first grid represents the object with the highest probability among the probabilities of each of the plurality of objects.
  • a variety of objects include trees, parked vehicles, buildings and other objects.
  • Trees correspond to a gray value of 1
  • parked vehicles correspond to a gray value of 2
  • buildings correspond to a gray value of 3
  • other objects correspond to a gray value of Take gray value 4 as an example, if the probability of the first grid representing trees is 0.95, the probability of representing parked vehicles is 0.02, the probability of representing buildings is 0.01, and the probability of representing other objects is 0.02, then the first grid represents The object with the highest probability among the various objects is a tree, so the gray value of the midpoint of the first grid is 1 corresponding to the tree.
  • the height information of the first grid may be used to indicate the height of points included in the first grid.
  • the height information of the first grid includes an average value of the heights of all points included in the first grid.
  • the average of the heights of all points included in the first grid z a is the height of the ath point among all points included in the first grid
  • N is the number of all points included in the first grid.
  • the height information of the first grid may be presented in the first grid through the color attribute parameters of the points in the first grid.
  • the color attribute parameter including lightness as an example, the height can be divided into multiple height intervals. Any two height intervals include different height values. Each height interval corresponds to a lightness. Then, the lightness corresponding to the point in the first grid It is the lightness corresponding to the height interval to which the height of the point belongs.
  • the reflection intensity information of the first grid may be used to indicate the reflection intensity of the points included in the first grid, for example, the reflectance of the object surface to which the points belong.
  • the reflection intensity information of the first grid includes an average value of reflection intensities of all points included in the first grid.
  • the average value of the reflection intensity of all points included in the first grid A a is the reflection intensity of the ath point among all the points included in the first grid.
  • the reflection intensity information of the first grid may be presented in the first grid through color attribute parameters of points in the first grid.
  • the color attribute parameter including hue as an example, the reflection intensity can be divided into multiple reflection intensity intervals. The reflection intensity values included in any two reflection intensity intervals are different, and each reflection intensity interval corresponds to a hue. Then, in the first grid The hue corresponding to the point is the hue corresponding to the reflection intensity interval to which the reflection intensity of the point belongs.
  • the positioning device may acquire the grid map corresponding to the first environment in at least the following two exemplary ways.
  • the positioning device obtains point cloud information corresponding to the first environment, and obtains a grid map corresponding to the first environment according to the point cloud information corresponding to the first environment.
  • the point cloud information corresponding to the first environment may be used to indicate position information of points in the point cloud corresponding to the first environment.
  • the information of the point cloud corresponding to the first environment may also be used to indicate the reflection intensity of the points in the point cloud corresponding to the first environment. The specific process of mode 1 will be described later in the method shown in FIG. 10 , and will not be repeated here.
  • the positioning device acquires the grid map corresponding to the first environment locally.
  • the grid map corresponding to the first environment is stored locally.
  • the positioning device obtains the grid map from other devices, it stores it locally.
  • the positioning device obtains the grid map according to the above-mentioned method 1, it stores it locally. Subsequently, when the grid map needs to be used, the positioning device acquires the grid map corresponding to the first environment locally.
  • the positioning device acquires a local grid corresponding to the second environment.
  • the second environment is included in the first environment, that is to say, the second environment is a part of the first environment.
  • the second environment is the surrounding environment where the device to be positioned is currently located, that is, the device to be positioned is in the second environment.
  • the second environment may be the north side of Building No. 2 in the residential area; taking the first environment as an industrial park as an example, the second environment may be an office building in the industrial park In front of the door; the first environment is the park, and the second environment can be under the rockery in the park.
  • the aforementioned partial grid may include at least one second grid.
  • the second grids may be square or rectangular, or at least one second grid has a part of the grids that are square and another part of the grids that are rectangular, which is not limited.
  • the size of the second grid can be fixed, that is, it is pre-stored in the positioning device, or it can be set by the user according to needs.
  • the local grid may be as shown in FIG. 4 .
  • the partial grid 402 in FIG. 4 includes 9 second grids.
  • the second grid in partial grid 402 is a square.
  • any one of the at least one second grid includes multiple points and attribute information of the second grid.
  • the attribute information of the second grid may include semantic information of the second grid and height information of the second grid.
  • the attribute information of the second grid may further include reflection intensity information of the second grid.
  • the semantic information of the second grid may be used to indicate the probability that the second grid represents each object in the plurality of objects.
  • the second grid represents the probability of each of the various objects, which can be understood as the object to which the point in the second grid belongs (or the object corresponding to the point in the second grid) is each of the various objects probability of an object.
  • the height information of the second grid may be used to indicate the height of points included in the second grid.
  • the reflection intensity information of the second grid may be used to indicate the reflection intensity of the points included in the second grid. It can be understood that the content included in the attribute information of the second grid is similar to the content included in the attribute information of the first grid. Therefore, the introduction of the attribute information of the second grid can refer to the attribute information of the first grid in S301 above. The description of the information will not be repeated here.
  • the positioning device acquires point cloud information corresponding to the second environment, and acquires a local grid corresponding to the second environment according to the point cloud information corresponding to the second environment.
  • the point cloud information corresponding to the second environment may be used to indicate position information of points in the point cloud corresponding to the second environment.
  • the point cloud information corresponding to the second environment may also be used to indicate the reflection intensity of points in the point cloud corresponding to the second environment.
  • the positioning device acquires a first matching probability.
  • the first matching probability is the matching probability of the local grid and the grid in the grid map when the local grid is located at different positions in the grid map.
  • the positioning device may determine the first matching probability corresponding to each type of attribute information based on each type of attribute information.
  • the positioning device acquires the first matching probability corresponding to each type of attribute information at different locations according to the attribute information of the grid in the local grid and the attribute information of the grid in the grid map. Specifically, when the positioning device acquires that the local grid is located in the first grid in the grid map, the first attribute information of the second grid in the local grid corresponds to the first attribute information of the first grid corresponding to the second grid. A first deviation between the attribute information. According to the first deviation, a first matching probability corresponding to the first attribute information of the first grid in the grid map is obtained.
  • the first attribute information of the second grid is any attribute information in the attribute information of the second grid.
  • the first attribute information of the second grid is semantic information of the second grid, or height information of the second grid, or reflection intensity information of the second grid.
  • the first attribute information of the first grid is the same as the first attribute information of the second grid. For example, if the first attribute information of the second grid is the semantic information of the second grid, then the first attribute information of the first grid is the semantic information of the first grid; if the first attribute information of the second grid is the height information of the second grid, then the first attribute information of the first grid is the height information of the first grid; if the first attribute information of the second grid is the reflection intensity information of the second grid, then The first attribute information of a grid is the reflection intensity information of the first grid.
  • the positioning device may also obtain the first predicted probability corresponding to each attribute information of the first grid in the grid map, so that the positioning device may obtain the corresponding The first matching probability and the first predicted probability determine the first posterior probability corresponding to each attribute information of the first grid, and then determine the undetermined probability according to the first posterior probability corresponding to each attribute information of the first grid position of the bit device.
  • the first predicted probability corresponding to the first attribute information of any first grid in the grid map is the probability that the local grid is currently located in the first grid predicted according to the first attribute information.
  • the first posterior probability corresponding to the first attribute information of any first grid in the grid map is the first predicted probability corresponding to the first attribute information of the first grid and the first attribute information of the first grid The product of the first matching probabilities corresponding to the information.
  • the following describes respectively when the first attribute information is semantic information, height information and reflection intensity information.
  • the positioning device acquires the first matching probability and the first prediction probability corresponding to the first attribute information of each first grid when the local grid is located in each first grid in the grid map. Specifically, reference may be made to the descriptions in Example 1-Example 3 below.
  • Example 1 When the first attribute information is semantic information, the positioning device obtains the first matching probability and The first process of predicting probabilities.
  • the positioning device can make the centers of the local grid 402 be located at the centers of the 63 first grids in the grid map 401 one by one, so as to obtain the local grid The first deviation between the semantic information of the nine second grids in 402 and the semantic information of the nine first grids corresponding to the nine second grids.
  • the partial grid 402 is equivalent to overlapping with the grid of the shaded part in the grid map 401 .
  • the first grid corresponding to the second grid (1,1) in the local grid 402 is the first grid (2,7) in the grid map 401
  • the second grid (1,2 ) is the first grid (2,8) in the grid map 401
  • the first grid corresponding to the second grid (1,3) in the partial grid 402 is the grid map 401
  • the first grid (2,9) of the local grid 402 corresponding to the second grid (2,1) is the first grid (3,7) in the grid map 401
  • the first grid corresponding to the second grid (2,2) in grid 402 is the first grid (3,8) in grid map 401 , and so on.
  • the second grid (a, b) represents the grid of row a and column b in the partial grid 402
  • the first grid (a, b) represents row a in the
  • the first deviation of the semantic information of the second grid (1,1) and the semantic information of the first grid (2,7) satisfy the formula Among them, s 1 is the first deviation between the semantic information of the second grid (1,1) and the semantic information of the first grid (2,7), n is the number of various objects in S301, and p b is the first deviation of the semantic information of the first grid (2,7).
  • the second grid represents the probability of the second object among the plurality of objects, represents the probability of the second object among the plurality of objects for the first grid.
  • the second object is any one of various objects.
  • the probability that the second grid represents trees is 0.2
  • the probability that it represents buildings is 0.85
  • the probability that it represents other objects is 0.05
  • the probability that the first grid represents trees is 0.95, indicating that the probability of a building is 0.01, and that of other objects is 0.04
  • s 1 (0.2-0.95)+(0.85-0.01)+(0.05-0.04).
  • the positioning device can obtain the semantic information of the second grid (1,2) and the first deviation s 2 of the semantic information of the first grid (2,8), and obtain the second grid ( 1,3) and the first deviation s 3 of the semantic information of the first grid (2,9), obtain the semantic information of the second grid (2,1) and the first grid (3,7)
  • the first deviation s 4 of the semantic information of the second grid (3,3) and the first deviation s 9 of the semantic information of the first grid (4,9) are acquired.
  • the positioning device may obtain the first total deviation between the semantic information of the second grid and the semantic information of the first grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401, that is, the first total deviation between the semantic information of the second grid and the semantic information of the first grid. a total deviation
  • the positioning device may obtain the first matching probability corresponding to the semantic information of the first black grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401 according to the first total deviation.
  • xi , y i ,m) may satisfy the formula: p(z
  • xi ,y i ,m) ⁇ e ⁇ S .
  • x i is the abscissa of the center position of the first black grid in the grid map 401
  • y i is the ordinate of the center position of the first black grid in the grid map 401 .
  • ⁇ is a constant.
  • the positioning device can use the same method as the above method to obtain when the center of the local grid 402 is located at the center of each of the other 62 first grids in the grid map 401, the second grid The first total deviation between the semantic information of the grid and the semantic information of the first grid, and according to each first total deviation, obtain each of the other 62 first grids whose center of the local grid 402 is located in the grid map 401 When the centers of the first grids, the semantic information of each first grid corresponds to the first matching probability.
  • the first prediction probability corresponding to the semantic information of any first grid in the grid map is a preset initial value, such as The ratio of 1 to the number of the first grid in the grid map, that is, the first prediction probability corresponding to the semantic information of any first grid satisfies the formula in, is the first prediction probability corresponding to the semantic information of any first grid, x i is the abscissa of the center of the first grid, y i is the ordinate of the center of the first grid, and num is the grid map The number of first rasters in .
  • the positioning device is not determining the position of the device to be positioned for the first time, the first prediction probability corresponding to the semantic information of any first grid in the grid map satisfies the formula:
  • n is a constant.
  • z,m) is the first posterior probability corresponding to the semantic information obtained last time when the position of the device to be positioned on the grid map is (x 0 ,y 0 ), (x 0 ,y 0 ) is the position of the device to be positioned in the grid map at the current moment predicted by the positioning device based on the position information of the device to be positioned at the previous moment and the motion parameters of the device to be positioned. According to the position information of the device to be positioned at the previous moment and the motion parameters of the device to be positioned, the positioning device can predict the position of the device to be positioned in the grid map at the current moment. Refer to the subsequent method shown in Figure 5 mentioned, and will not be repeated here.
  • ⁇ is the variance.
  • the center of the local grid can also be replaced by other positions, such as the upper left corner of the local grid (such as the upper left corner of the second grid (1,1) in the local grid 402), the local grid upper right corner (such as the upper right corner of the second grid (1,3) in the local grid 402), the lower left corner of the local grid (such as the lower left corner of the second grid (3,1) in the local grid 402), The lower right corner of the local grid (such as the upper right corner of the second grid (3,3) in the local grid 402 ), etc.
  • the center of the first grid can also be replaced with other positions.
  • the center of the first grid can be replaced by the upper left corner of the first grid (such as the first grid (2,7 )); if the center of the local grid is replaced by the upper right corner of the local grid, then the center of the first grid can be replaced by the upper right corner of the first grid (such as the first grid in the grid map 401 (2,7) at the upper right corner); if the center of the local grid is replaced with the lower left corner of the local grid, then the center of the first grid can be replaced with the lower left corner of the first grid (as in the grid map 401 The lower left corner of the first grid (2,7)); if the center of the local grid is replaced by the lower right corner of the local grid, the center of the first grid can be replaced by the lower right corner of the first grid (such as grid Lower right corner of first grid (2,7) in map 401).
  • Example 2 When the first attribute information is height information, the positioning device acquires the first matching probability and The first process of predicting probabilities.
  • the semantic information of the nine second grids in the local grid 402 obtained by the positioning device corresponds to the nine second grids respectively.
  • the first deviation between the semantic information of a grid is similar, and the positioning device can obtain the center of the local grid 402 when the center of the local grid 402 is located at the center of the 63 first grids in the grid map 401 one by one.
  • the positioning device makes the center of the partial grid 402 located at the center of the first black grid in the grid map 401 , then the partial grid 402 is equivalent to overlapping with the grid of the shaded part in the grid map 401 .
  • the first deviation between the height information of the second grid (1,1) and the height information of the first grid (2,7) Satisfies the formula: in, is the first deviation between the height information of the second grid (1,1) and the height information of the first grid (2,7), h 1 is the height of all points included in the second grid (1,1) average value, is the average height of all points included in the first grid (2,7).
  • the positioning device can obtain the first deviation between the height information of the second grid (1, 2) and the height information of the first grid (2, 8) Obtain the first deviation between the height information of the second grid (1,3) and the height information of the first grid (2,9) Obtain the first deviation between the height information of the second grid (2,1) and the height information of the first grid (3,7) Obtain the first deviation between the height information of the second grid (3,3) and the height information of the first grid (4,9) Subsequently, the positioning device can obtain the first total deviation between the height information of the second grid and the height information of the first grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401, that is, the first total deviation a total deviation Subsequently, the positioning device may obtain the first matching probability corresponding to the height information of the first black grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401 according to the first total deviation.
  • the positioning device can use the same method as the above method to obtain when the center of the local grid 402 is located at the center of each of the other 62 first grids in the grid map 401, the second grid
  • the height information of each first grid corresponds to the first matching probability.
  • the first prediction probability corresponding to the height information of any first grid in the grid map is a preset initial value, such as The ratio of 1 to the number of first grids in the grid map, that is, the first predicted probability corresponding to the height information of any first grid satisfies the formula in, is the first prediction probability corresponding to the height information of any first grid, x i is the abscissa of the center of the first grid, and y i is the ordinate of the center of the first grid.
  • the positioning device is not determining the position of the device to be positioned for the first time, the first prediction probability corresponding to the height information of any first grid in the grid map satisfies the formula:
  • n is a constant.
  • z,m) is the first posterior probability corresponding to the height information obtained last time when the position of the device to be positioned on the grid map is (x 0 ,y 0 ), (x 0 ,y 0 ) is the position of the device to be positioned in the grid map at the current moment predicted by the positioning device based on the position information of the device to be positioned at the previous moment and the motion parameters of the device to be positioned.
  • ⁇ is the variance.
  • Example 3 When the first attribute information is reflection intensity information, the positioning device acquires the first matching probability corresponding to the first attribute information of each first grid when the local grid is located in each first grid in the grid map and the first process of predicting probabilities.
  • the process of the positioning device obtaining the first matching probability corresponding to the first attribute information of each first grid is different from that in the case where the first attribute information is height information , the process of the positioning device acquiring the first matching probability corresponding to the first attribute information of each first grid is similar.
  • the positioning device can obtain the local The first deviation between the reflection intensity information of the nine second grids in the grid 402 and the reflection intensity information of the nine first grids corresponding to the nine second grids respectively.
  • the partial grid 402 is equivalent to overlapping with the grid of the shaded part in the grid map 401 .
  • the reflection intensity information of the second grid (1,1) and the reflection intensity information of the first grid (2,7) A deviation can satisfy the formula: in, is the first deviation between the reflection intensity information of the second grid (1,1) and the reflection intensity information of the first grid (2,7), and a 1 is the value of all points included in the second grid (1,1)
  • the mean value of the reflected intensity is the average value of reflection intensities of all points included in the first grid (2,7).
  • the positioning device can obtain the first deviation between the reflection intensity information of the second grid (1, 2) and the reflection intensity information of the first grid (2, 8) Obtain the first deviation between the reflection intensity information of the second grid (1,3) and the reflection intensity information of the first grid (2,9) Obtain the first deviation between the reflection intensity information of the second grid (2,1) and the reflection intensity information of the first grid (3,7) Obtain the first deviation between the reflection intensity information of the second grid (3,3) and the reflection intensity information of the first grid (4,9) Subsequently, the positioning device may obtain the first total deviation between the reflection intensity information of the second grid and the reflection intensity information of the first grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401, the first total deviation Subsequently, the positioning device may obtain the first matching probability corresponding to the reflection intensity information of the first black grid when the center of the local grid 402 is located at the center of the first black grid in the grid map 401 according to the first total deviation.
  • the first matching probability corresponding to the reflection intensity
  • the positioning device can use the same method as the above method to obtain when the center of the local grid 402 is located at the center of each of the other 62 first grids in the grid map 401, the second grid The first total deviation between the reflection intensity information of the grid and the reflection intensity information of the first grid, according to each first total deviation, the center of the local grid 402 is located in the other 62 first grids in the grid map 401 When the center of each first grid of , the first matching probability corresponding to the reflection intensity information of each first grid.
  • the first prediction probability corresponding to the reflection intensity information of any first grid in the grid map is a preset initial value, For example, the ratio of 1 to the number of the first grid in the grid map, that is, the first predicted probability corresponding to the reflection intensity information of any first grid satisfies the formula in, is the first predicted probability corresponding to the reflection intensity information of any first grid, x i is the abscissa of the center of the first grid, and y i is the ordinate of the center of the first grid.
  • the positioning device is not determining the position of the device to be positioned for the first time, the first prediction probability corresponding to the reflection intensity information of any first grid in the grid map satisfies the formula:
  • n is a constant.
  • z,m) is the first posterior probability corresponding to the reflection intensity information obtained last time when the position of the device to be positioned on the grid map is (x 0 ,y 0 ), (x 0 , y 0 ) is the predicted position of the device to be positioned in the grid map at the current moment by the positioning device based on the position information of the device to be positioned at the previous moment and the motion parameters of the device to be positioned.
  • ⁇ is the variance.
  • the positioning device determines the position of the device to be positioned according to the first matching probability.
  • the position of the device to be located is the position of the device to be located in the grid map.
  • the position of the device to be positioned is the position of the centroid of the smart car in the grid map.
  • the positioning device determines the location information corresponding to each type of attribute information according to the first matching probability corresponding to each type of attribute information at different locations, and determines the location of the device to be positioned according to the location information corresponding to each type of attribute information.
  • the positioning device determines the first matching probability corresponding to each type of attribute information at different locations and the first predicted probability corresponding to each type of attribute information at different locations, and determines the first matching probability corresponding to each type of attribute information at different locations.
  • a posterior probability according to the first posterior probability corresponding to each attribute information at different locations, determine the location information corresponding to each attribute information.
  • the first posterior probability corresponding to the semantic information of any first grid in the grid map is that the first predicted probability corresponding to the semantic information of the first grid corresponds to the semantic information of the first grid The product of the first matching probabilities of .
  • the first posterior probability corresponding to the semantic information of the first black grid in the grid map 401 Satisfies the formula:
  • the location information corresponding to the semantic information satisfies the formula: in, is the abscissa of the device to be located in the grid map determined based on the semantic information, that is, the abscissa of the device to be located in the grid map corresponding to the semantic information
  • x i is the ith first grid in the grid map
  • z, m) is the first posterior probability corresponding to the semantic information of the i-th first grid in the grid map
  • N is included in the grid map the number of first grids, is the ordinate of the device to be located in the grid map determined based on the semantic information, that is, the ordinate of the device to be located
  • the first posterior probability corresponding to the height information of any first grid in the grid map is that the first predicted probability corresponding to the height information of the first grid corresponds to the height information of the first grid The product of the first matching probabilities of .
  • the first posterior probability corresponding to the height information of the first black grid in the grid map 401 Satisfies the formula:
  • the location information corresponding to the height information satisfies the formula: in, is the abscissa of the device to be located in the grid map determined based on the height information, that is, the abscissa of the device to be located in the grid map corresponding to the height information, q( xi ,y i
  • the first posterior probability corresponding to the height information of the i-th first grid in the map is the vertical coordinate of the device to be positioned in the grid map determined based on the height information, that is, the vertical coordinate of the device to be positioned corresponding to the height information in the grid map.
  • the first posterior probability corresponding to the reflection intensity information of any first grid in the grid map is the first predicted probability corresponding to the reflection intensity information of the first grid and the first prediction probability of the first grid The product of the first matching probabilities corresponding to the reflection intensity information.
  • the reflection intensity information of the first black grid in the grid map 401 corresponds to the first posterior
  • the probability satisfies the formula:
  • the position information corresponding to the reflection intensity information satisfies the formula: in, is the abscissa of the device to be located in the grid map determined based on the reflection intensity information, that is, the abscissa of the device to be located in the grid map corresponding to the reflection intensity information, r(xi , y i
  • the following describes the process by which the positioning device determines the position of the device to be positioned according to the position information corresponding to each type of attribute information.
  • the positioning device weights and sums the position information corresponding to each type of attribute information to obtain the position of the device to be positioned.
  • the weight corresponding to each attribute information may be determined first. It can be understood that the weights corresponding to each type of attribute information may be the same or different, for example, the weights corresponding to each type of attribute information are 1.
  • the weight corresponding to each attribute information may be preset and stored in the positioning device, or may be determined according to a certain algorithm.
  • the weight corresponding to each attribute information is determined according to the distribution of the first matching probability corresponding to the attribute information. For example, the more concentrated the distribution of the first matching probability corresponding to the attribute information in the grid map is, the higher the weight corresponding to the attribute information is, and the less concentrated the distribution of the first matching probability corresponding to the attribute information in the grid map is, the more concentrated the distribution of the attribute information is. The lower the weight corresponding to the information.
  • the weight corresponding to each attribute information is determined according to the variance of the first matching probability corresponding to the attribute information.
  • the weight corresponding to semantic information can satisfy the formula: Among them, w p is the weight corresponding to the semantic information, ⁇ is a constant, and up is the variance of the first matching probability corresponding to the semantic information.
  • the weight corresponding to the height information can satisfy the formula: Wherein, w h is the weight corresponding to the height information, and u h is the variance of the first matching probability corresponding to the height information.
  • the weight corresponding to the reflection intensity information can satisfy the formula: Wherein, w a is the weight corresponding to the reflection intensity information, and u a is the variance of the first matching probability corresponding to the reflection intensity information. It can be understood that the smaller the variance, the more concentrated the distribution of the first matching probability corresponding to the variance is, and the larger the variance is, the less concentrated the distribution of the first matching probability is corresponding to the variance.
  • the position of the device to be positioned satisfies the following formula: Wherein, X is the abscissa of the device to be positioned in the grid map determined by the positioning device, and Y is the vertical coordinate of the device to be positioned in the grid map determined by the positioning device.
  • the positioning device determines the location information corresponding to each attribute information, it can first screen the location information to eliminate inaccurate location information, Then the position of the device to be positioned is determined according to the filtered position information.
  • the positioning device determines the location information corresponding to at least one attribute information that meets the preset rules in the location information corresponding to each attribute information, and determines the location information of the device to be positioned according to the location information corresponding to at least one attribute information s position.
  • the preset rules may include at least one of the following rules: Among the location information corresponding to each attribute information, the location information whose distance from the first location is less than or equal to the first value is determined as the location corresponding to at least one attribute information information; or, among the location information corresponding to each attribute information, the location information whose matching degree is greater than or equal to the second value is determined as the location information corresponding to at least one attribute information.
  • the first position is the position of the device to be positioned in the grid map at the current moment predicted by the positioning device according to the position information of the device to be positioned at a previous moment and the motion parameters of the device to be positioned.
  • the matching degree of the position information corresponding to any kind of attribute information can be based on the ICP algorithm, and the point corresponding to the position information in the point cloud map corresponding to the first environment is matched with the point in the point cloud corresponding to the first environment owned.
  • the location information corresponding to at least one attribute information is determined as the location of the device to be positioned.
  • the method of determining the weight corresponding to the attribute information is as described above, and will not be repeated here.
  • the positioning device may also not determine the first posterior probability corresponding to each type of attribute information at different locations, but directly determines the corresponding The location information of each attribute information, and then determine the location of the device to be located according to the location information corresponding to each attribute information.
  • the first posterior probability corresponding to the attribute information takes into account the first predicted probability corresponding to the attribute information, therefore, the position of the device to be positioned determined according to the first posterior probability corresponding to each attribute information in different positions is more than It is more accurate to determine the position of the device to be located directly according to the first matching probability corresponding to each attribute information of different positions.
  • the positioning device can obtain the grid map corresponding to the first environment, and obtain the local grid corresponding to the second environment information, and according to the various attribute information of the first grid in the grid map and the local grid Multiple attribute information of the second grid can locate the positioning device, so that the positioning device does not only rely on a certain attribute for positioning, thereby improving the accuracy and stability of positioning.
  • the positioning device acquires the first matching probabilities of the local grids and the grids in the grid map one by one at each positioning, and the calculation cost is relatively large.
  • the positioning device may obtain the matching probabilities of the local grids and the grids in the grid map one by one.
  • the positioning device may match the local grid with the grid in the grid map based on a histogram filter method.
  • the positioning device may first predict the position of the device to be positioned on the grid map, and then obtain the matching probability between the local grid and the grid corresponding to the position in the grid map. In this way, the matching probabilities between the local grids and the grids in the grid map do not need to be obtained one by one, which reduces the calculation cost.
  • the method shown in FIG. 3 may further include S305.
  • the positioning device predicts the first area corresponding to the position of the device to be positioned at the second moment according to the pose information of the device to be positioned at the first moment and the motion parameters of the device to be positioned.
  • the first moment is the last time when the positioning device performs positioning for the device to be positioned.
  • the pose information at the first moment is used to indicate the latest position and orientation angle of the device to be positioned in the grid map determined by the positioning device.
  • the orientation angle of the device to be located on the grid map may indicate the direction of movement of the device to be located on the grid map.
  • the second moment is the current moment.
  • the position at the second moment is the position of the device to be located in the grid map at the current moment. That is to say, the positioning device can predict the position of the device to be positioned at the current moment according to the latest determined pose information of the device to be positioned and the motion parameters of the device to be positioned.
  • the motion parameter may be a parameter acquired by the perception module in real time.
  • the sensing module may be the sensing device 102 in FIG. 1A , or the sensing module 1112 in FIG. 1B .
  • the motion parameters are inertial measurement unit (IMU) data.
  • the sensing module can also be a wheel speed sensor installed on the device to be positioned, in this case, the motion parameter is a wheel speed meter parameter.
  • the positioning device acquires the motion parameters, parameters such as the angular velocity of the device to be positioned and the linear velocity of the device to be positioned can be obtained according to the motion parameters.
  • the positioning device can also obtain the rotation angle of the wheel of the device to be positioned according to the motion parameters.
  • the angular velocity of the device to be positioned, the linear velocity of the device to be positioned and the rotation angle of the wheel of the device to be positioned may be as shown in FIG. 6 .
  • Lines 601 and 602 in FIG. 6 are used to represent the two rear wheels of the smart car, and lines 603 and 604 are used to represent the two front wheels of the smart car.
  • the angular velocity ⁇ of the device to be positioned may be the arc traversed by the device to be positioned per unit time.
  • the linear velocity v of the device to be positioned can be the distance traveled by the center point of the line connecting the two rear wheels of the smart car in unit time.
  • the rotation angle ⁇ of the wheel of the device to be positioned may be the angle between the wheel of the smart car and the line connecting the center of the front wheel and the center of the rear wheel of the smart car.
  • the pose (x t1 , y t1 , ⁇ t1 ) of the device to be positioned at the first moment and the predicted position (x t2 , y t2 ) of the device to be positioned at the second moment satisfy the formula:
  • x t1 is the abscissa of the device to be positioned in the grid map at the first moment.
  • y t1 is the vertical coordinate of the device to be positioned in the grid map at the first moment.
  • x t2 is the predicted abscissa of the device to be positioned in the grid map at the second moment.
  • y t2 is the predicted vertical coordinate of the device to be positioned in the grid map at the second moment.
  • v is the linear velocity of the device to be positioned.
  • ⁇ is the angular velocity of the device to be positioned.
  • ⁇ t is the difference between the second moment and the first moment.
  • the positioning device predicts the position of the device to be positioned at the second moment. Therefore, after the positioning device predicts the position of the device to be positioned at the second moment, according to the predicted position of the device to be positioned at the second moment The position at the second moment determines the first area. Subsequently, the positioning device may acquire a first matching probability between the local grid and the grid in the first area, and determine the position of the device to be positioned according to the first matching probability. In this way, the positioning accuracy and stability of the positioning device can be improved, and the calculation cost can also be reduced.
  • the positioning device may determine an area within a certain range centered on the predicted position of the device to be positioned at the second moment as the first area. For example, if the pose of the device to be positioned at the first moment and the predicted position of the device to be positioned at the second moment are shown in FIG. 7 , the first area may be the shaded part in FIG. 7 .
  • the positioning device may not need to obtain the first matching probability according to the method shown in FIG. 3 above, and determine the position of the device to be positioned according to the first matching probability.
  • the positioning device may obtain the first matching probability of the local grid and the grid in the previous map when the local grid is located in the first area, that is, the first matching probability of the grid in the first area, according to the first matching probability of the grid in the first area.
  • a first matching probability of grids in an area determines the location of the device to be located. The details are described below.
  • the positioning device acquires the first matching probability corresponding to each type of attribute information at different positions according to the attribute information of the grids in the local grids and the attribute information of the grids in the first region. Specifically, when the positioning device acquires the first grid in which the local grid is located in the first area, the first attribute information of the second grid in the local grid corresponds to the first attribute information of the first grid corresponding to the second grid. A first deviation between the attribute information. According to the first deviation, a first matching probability corresponding to the first attribute information of the first grid in the first region is acquired.
  • the first attribute information of the second grid is any attribute information in the attribute information of the second grid, and the first attribute information of the first grid has the same attribute as the first attribute information of the second grid.
  • the positioning device may also acquire the first predicted probability corresponding to each attribute information of the first grid in the first area, so that the positioning device may The first matching probability and the first prediction probability corresponding to the information determine the first posterior probability corresponding to each attribute information of the first grid in the first area, and then according to each type of attribute information of the first grid in the first area The first posterior probability corresponding to the attribute information determines the position of the device to be located.
  • the process of the positioning device obtaining the first predicted probability corresponding to each attribute information of the first grid in the first area is the same as the positioning device obtaining the first prediction probability corresponding to each attribute information of the first grid in the grid map in S303 above.
  • the process of predicting the probability is similar, so reference can be made to the corresponding description in S303 above, and details are not repeated here.
  • the positioning device determines the location information corresponding to each type of attribute information according to the first matching probability corresponding to each type of attribute information of the first grid in the first area, and according to the location information corresponding to each type of attribute information, Determine the location of the device to be located. Specifically, according to the first matching probability and the first prediction probability corresponding to each attribute information of the first grid in the first area, the positioning device determines the first matching probability corresponding to each attribute information of the first grid in the first area. A posterior probability, according to the first posterior probability corresponding to each attribute information of the first grid in the first area, determine the location information corresponding to each attribute information, and determine the undetermined location information according to the location information corresponding to each attribute information position of the bit device.
  • the above process is similar to the process of determining the position of the device to be positioned by the positioning device in S304 above, so reference may be made to the corresponding description in S304 above, and details are not repeated here.
  • the positioning device determines the location information corresponding to each attribute information, it can first screen the location information to eliminate inaccurate location information, Then the position of the device to be positioned is determined according to the filtered position information. Specifically, reference may be made to the corresponding description in S304 above, and details are not repeated here.
  • the positioning device can locate the device to be positioned and acquire the position of the device to be positioned.
  • the positioning device may also obtain the orientation angle of the device to be positioned.
  • the method shown in FIG. 3 further includes S306-S307.
  • S306 The positioning device acquires a second matching probability.
  • the second matching probability is the matching probability between the partial grid and the grid in the grid map when the partial grid is located at the position of the device to be positioned and the partial grid is rotated by different angles. Different rotation angles of the local grid correspond to different orientation angles of the device to be positioned.
  • the center of the local grid can be located at the position of the device to be positioned, such as the local grid 901 in Figure 9, and the local grid can be rotated around the position of the device to be positioned to obtain different rotations.
  • the local grids of the corners such as the local grids 902-904 in FIG. 9 .
  • the orientation angle of the device to be positioned corresponding to the rotation angle of the partial grid 901 is the orientation angle corresponding to the arrow 905
  • the orientation angle of the device to be positioned corresponding to the rotation angle of the partial grid 902 is the orientation angle corresponding to the arrow 906
  • the partial grid 903 The orientation angle of the device to be positioned corresponding to the rotation angle of is the orientation angle corresponding to the arrow 907
  • the orientation angle of the device to be positioned corresponding to the rotation angle of the partial grid 904 is the orientation angle corresponding to the arrow 908 .
  • the positioning device can acquire the second matching probability between the local grid and the grid in the grid map at different rotation angles, for example, the second matching probability between the local grid 901 and the grid in the grid map, the local grid 902 the second matching probability with the grid in the grid map, the partial grid 903 the second matching probability with the grid in the grid map, the partial grid 904 the second matching probability with the grid in the grid map .
  • the rotation angle of the local grid is the angle between the rotated local grid pattern and the local grid initially located at the position of the device to be positioned.
  • the rotation angle of the local grid 902 is the angle between the local grid 902 and the local grid 901
  • the rotation angle of the local grid 903 is the angle between the local grid 903 and the local grid 901. angle.
  • the difference between the rotation angles of two adjacent rotations of the local grid may be the same or different, which is not limited.
  • the positioning device acquires the second matching probabilities of multiple rotation angles corresponding to the local grid when the local grid is located at the position of the device to be positioned.
  • the center of the partial grid is located at the position of the device to be positioned, and the partial grid is rotated around the position of the device to be positioned to obtain multiple rotation angles.
  • the plurality of rotation angles may be 0°, 45°, 90°, 135°, 180°, 225°, 270° and 315°.
  • the positioning device can first predict the orientation angle of the device to be positioned according to the orientation angle of the device to be positioned at the third moment and the motion parameters of the device to be positioned.
  • the orientation angle of the positioning device at the second moment, and a plurality of rotation angles are determined according to the orientation angle at the second moment. In this case, it is not necessary to rotate the local grid around the position of the device to be positioned to obtain multiple rotation angles. Instead, the local grid is centered on the position of the device to be positioned, and at the second moment Rotate around the orientation angle to get multiple rotation angles.
  • the included angle between each rotation angle and the orientation angle at the second moment is smaller than or equal to the first angle. In this way, the calculation overhead of the positioning device can be reduced, and the accuracy of the determined orientation angle of the device to be positioned can be improved.
  • the third moment is the last time when the positioning device determines the orientation angle for the device to be positioned.
  • the orientation angle at the third moment is the orientation angle of the device to be located in the grid map determined by the positioning device last time.
  • the third moment and the first moment may be the same or different.
  • the orientation angle of the device to be located on the grid map may indicate the direction of movement of the device to be located on the grid map.
  • the second moment is the current moment.
  • the orientation angle at the second moment is the orientation angle of the device to be located in the grid map at the current moment. That is to say, the positioning device can predict the orientation angle of the device to be located on the grid map at the current moment according to the latest determined orientation angle of the device to be located and the motion parameters of the device to be located.
  • the motion parameters For the introduction of the motion parameters, reference may be made to the corresponding description in S305 above, which will not be repeated here.
  • parameters such as the angular velocity of the device to be positioned may be obtained according to the motion parameters.
  • the positioning device can also obtain the linear velocity of the device to be positioned, the rotation angle of the wheel of the device to be positioned, etc. according to the motion parameters.
  • ⁇ is the angular velocity of the device to be positioned.
  • ⁇ t is the difference between the second moment and the third moment.
  • the positioning device may determine the second matching probability corresponding to each type of attribute information based on each type of attribute information. Exemplarily, according to the attribute information of the grid in the local grid and the attribute information of the grid in the grid map, the positioning device acquires the second matching probability corresponding to each attribute information of the local grid corresponding to the multiple rotation angles .
  • the positioning device obtains the information corresponding to each type of attribute information of the first rotation angle according to the attribute information of the grid in the local grid corresponding to the first rotation angle and the attribute information of the grid in the grid map.
  • Second match probability wherein, the first rotation angle is any rotation angle among multiple rotation angles, and the second matching probability corresponding to each attribute information of the first rotation angle is each type of the grid in the local grid corresponding to the first rotation angle. Attribute information, the matching probability of each attribute information of the raster in the raster map.
  • the positioning device obtains the local grid at the position of the device to be positioned, and when the local grid is rotated by the first rotation angle, the first attribute information of each second grid in the local grid, and the second grid According to the second deviation between the first attribute information of the first grid corresponding to the grid, the second matching probability corresponding to the first attribute information of the first rotation angle is obtained according to the second deviation.
  • the first attribute information of the second grid is any attribute information in the attribute information of the second grid, and the first attribute information of the first grid has the same attribute as the first attribute information of the second grid.
  • the process that the positioning device obtains the second deviation, and obtains the second matching probability corresponding to the first attribute information of the first rotation angle according to the second deviation is the same as the above S303 in which the positioning device obtains the first deviation, and obtains the second matching probability according to the first deviation.
  • the process of the first matching probability corresponding to the first attribute information of the first grid in the grid map is similar, and reference may be made to the corresponding description in S303 above, and details are not repeated here.
  • the positioning device can also obtain the second predicted probability corresponding to each attribute information of the first rotation angle, so that the positioning device can use the second matching probability and the second prediction probability corresponding to each attribute information of the first rotation angle.
  • the prediction probability is to determine the second posterior probability corresponding to each attribute information of the first rotation angle, and then determine the orientation angle of the device to be positioned according to the second posterior probability corresponding to each attribute information of the first rotation angle.
  • the second prediction probability corresponding to each attribute information of the first rotation angle is a preset initial value, such as 1/360.
  • the positioning device is not determining the position of the device to be positioned for the first time, the second prediction probability corresponding to the first attribute information of the first rotation angle satisfies the formula:
  • ⁇ is the second prediction probability corresponding to the attribute information of the first rotation angle.
  • ⁇ is a constant.
  • z,m) is the second posterior probability corresponding to the first attribute information of the device to be positioned when the orientation angle of the grid map is ⁇ 0 acquired last time, and ⁇ 0 is the The orientation angle of the device at the last moment and the motion parameters of the device to be positioned, and the predicted orientation angle of the device to be positioned in the grid map at the current moment.
  • ⁇ i is the orientation angle of the device to be positioned corresponding to the first rotation angle.
  • ⁇ is the variance.
  • the positioning device determines the orientation angle of the device to be positioned according to the second matching probability.
  • the positioning device determines the orientation angle information corresponding to each attribute information according to the second matching probability corresponding to each attribute information of each rotation angle, and determines the undetermined orientation angle information according to the orientation angle information corresponding to each attribute information.
  • the orientation angle of the device determines the second posterior probability corresponding to each attribute information of each rotation angle according to the second matching probability and the second prediction probability corresponding to each attribute information of each rotation angle, and according to each rotation angle According to the second posterior probability corresponding to each attribute information, the orientation angle information corresponding to each attribute information is determined, and the orientation angle of the device to be positioned is determined according to the orientation angle information corresponding to each attribute information.
  • the second posterior probability corresponding to the first attribute information of the first rotation angle satisfies the formula: Among them, k( ⁇ i
  • the orientation angle information corresponding to the first attribute information satisfies the formula: in, is the orientation angle indicated by the orientation angle information corresponding to the first attribute information, that is, the orientation angle of the device to be positioned determined based on the first attribute information, ⁇ i is the orientation angle of the device to be positioned corresponding to the i-th rotation angle among the multiple rotation angles Orientation angle, j is the number of multiple rotation angles.
  • the positioning device weights and sums the orientation angle information corresponding to each type of attribute information to obtain the orientation angle of the device to be positioned.
  • the weights corresponding to each type of attribute information reference may be made to the corresponding description in S304 above, which will not be repeated here.
  • the orientation angle of the device to be positioned satisfies the following formula: Wherein, ⁇ is the orientation angle of the device to be positioned determined by the positioning device, w p is the weight corresponding to the semantic information, w h is the weight corresponding to the height information, and w a is the weight corresponding to the reflection intensity information. is the orientation angle of the device to be positioned based on the semantic information, is the orientation angle of the device to be positioned based on the height information, is the orientation angle of the device to be positioned based on the reflection intensity information.
  • the positioning device may also not determine the second posterior probability corresponding to each attribute information of multiple rotation angles, but directly determines each The orientation angle information corresponding to each type of attribute information, and then according to the orientation angle information corresponding to each type of attribute information, the orientation angle of the device to be positioned is determined.
  • the second posterior probability corresponding to the attribute information takes into account the second predicted probability corresponding to the attribute information, therefore, the orientation of the device to be positioned is determined according to the second posterior probability corresponding to each attribute information of multiple rotation angles The angle is more accurate than the orientation angle of the device to be positioned which is determined directly according to the second matching probability corresponding to each attribute information of the plurality of rotation angles.
  • S301 may include S3011 and S3012.
  • the positioning device acquires point cloud information corresponding to the first environment.
  • the point cloud information corresponding to the first environment may be used to indicate the location information of the points in the point cloud corresponding to the first environment, such as the coordinates of the points in the point cloud corresponding to the first environment in the three-dimensional map.
  • the three-dimensional map may be a point cloud map.
  • the point cloud information corresponding to the first environment is also used to indicate the reflection intensity of the points in the point cloud corresponding to the first environment.
  • the positioning device acquires point cloud information corresponding to the first environment through the sensing device.
  • the sensing device may or may not be deployed on the positioning device.
  • the laser radar can move in the first environment and emit laser light to collect point cloud information in the first environment.
  • the positioning device is based on SLAM algorithm, such as lightweight and ground optimized lidar odometry and mapping (Lego-Loam) algorithm, Obtain the point cloud map corresponding to the first environment.
  • the point cloud in the first environment is a point cloud obtained by reflecting the lidar from objects in the first environment.
  • the reflection intensity of a point in the point cloud corresponding to the first environment is the reflection intensity of the point to the laser.
  • point cloud information when collecting point cloud information through lidar, point cloud information can be collected once at a certain distance.
  • each acquired point cloud can be down-sampled by a voxel raster filter in the point cloud library.
  • the voxel grid filter can achieve the function of downsampling without destroying the geometric structure of the point cloud itself. After downsampling, the number of points in the point cloud collected each time can be reduced while maintaining the shape characteristics of the point cloud.
  • the positioning device acquires a grid map corresponding to the first environment according to the point cloud information corresponding to the first environment.
  • the positioning device projects (for example, projecting onto the ground) the point cloud map corresponding to the first environment to obtain a two-dimensional map corresponding to the first environment, and divides the two-dimensional map into a plurality of first grids, To obtain the grid map corresponding to the first environment.
  • the height of each point in the grid map is the value of the vertical coordinate (z coordinate) of the point in the point cloud map.
  • the reflection intensity of each point in the grid map is the reflection intensity of the point to the laser.
  • the semantic information of the first grid may be determined according to the semantic information of points in the first grid.
  • the semantic information of a point in the first grid is the probability that the object to which the point belongs (or the object corresponding to the point) is each of the various objects.
  • p m [i] is the probability that the first grid represents any one of multiple objects
  • ⁇ is a constant
  • count(label i ) 1 is the points in the first grid that belong to any one of the objects
  • M is the number of points in the first grid.
  • the point in the first grid is input into the semantic model, and the probability that the object to which the point belongs (or the object corresponding to the point) is each of various objects can be obtained.
  • the semantic model may be stored in the positioning device.
  • the semantic model may also be obtained by training the positioning device before S3012.
  • the positioning device can use PointNet for training based on various objects marked in advance to obtain a semantic model.
  • the actions of the positioning device in S3011-S3012 above can be executed by the processor 201 in the positioning device 20 shown in FIG. 2 calling the application program code stored in the memory 203, which is not limited in this embodiment of the present application.
  • S302 may include S3021 and S3022.
  • the positioning device acquires point cloud information corresponding to the second environment.
  • the point cloud information corresponding to the second environment may be used to indicate position information of points in the point cloud corresponding to the second environment, such as coordinates of points in the point cloud corresponding to the second environment in the three-dimensional map.
  • the three-dimensional map may be a point cloud map.
  • the point cloud information corresponding to the second environment is also used to indicate the reflection intensity of the points in the point cloud corresponding to the second environment.
  • a sensing device is deployed on the device to be positioned, and the point cloud information corresponding to the second environment is obtained through the sensing device.
  • the sensing device as a lidar
  • the device to be positioned can move in the first environment, and when it moves to the second environment, emit laser light through the lidar to collect point cloud information in the second environment. If the device to be positioned is the same as the positioning device, the device to be positioned obtains a point cloud map corresponding to the second environment based on a SLAM algorithm, such as a Lego-Loam algorithm.
  • the device to be positioned sends point cloud information in the second environment to the positioning device, and the positioning device obtains a point cloud map corresponding to the second environment based on a SLAM algorithm, such as a Lego-Loam algorithm.
  • the point cloud in the second environment is a point cloud obtained by reflecting the lidar from objects in the second environment.
  • the reflection intensity of a point in the point cloud corresponding to the second environment is the reflection intensity of the point to the laser.
  • point cloud information when collecting point cloud information through lidar, point cloud information can be collected once at a certain distance.
  • each acquired point cloud can be down-sampled by a voxel raster filter in the point cloud library.
  • the positioning device acquires a local grid corresponding to the second environment according to the point cloud information corresponding to the second environment.
  • the positioning device projects the point cloud map corresponding to the second environment (such as projecting onto the ground) to obtain a two-dimensional map corresponding to the second environment, divides the two-dimensional map into a plurality of second grids, To get the local grid corresponding to the second environment.
  • the height of each point in the local grid is the value of the vertical coordinate (z coordinate) of the point in the point cloud map.
  • the reflection intensity of each point in the local grid is the reflection intensity of the point to the laser.
  • the semantic information of the second grid may be determined according to the semantic information of points in the second grid.
  • the semantic information of a point in the second grid is the probability that the object to which the point belongs (or the object corresponding to the point) is each of the various objects.
  • p[i] is the probability that the second grid represents any one of the various objects
  • ⁇ is a constant
  • count(label i ) 2 is the number of points belonging to the arbitrary object in the second grid
  • W is the number of points in the second grid.
  • the point in the second grid is input into the semantic model, and the probability that the object to which the point belongs (or the object corresponding to the point) is each of various objects can be obtained.
  • the semantic model For the introduction of the semantic model, reference may be made to the corresponding description in S3012 above, and details are not repeated here.
  • the actions of the positioning device in S3021-S3022 above can be executed by the processor 201 in the positioning device 20 shown in FIG. 2 calling the application program code stored in the memory 203, which is not limited in this embodiment of the present application.
  • the methods and/or steps implemented by the positioning device may also be implemented by components (such as chips or circuits) that can be used in the positioning device, without limitation.
  • the positioning devices to be positioned in 4 positions were respectively positioned using method 1, method 2 and method 4, and 50 positioning results at different times were counted. After the statistics The positioning results are shown in Table 1.
  • Table 1 for position 1, that is, the actual position of the device to be positioned as shown in Figure 12, in the 50 positionings, the proportion of the positioning error of the method 1 when the device to be positioned is less than 10 cm is 58.9%, and the method 2 to be positioned is 58.9%.
  • the proportion of positioning error less than 10cm when the device is positioned is 68.1%, and the proportion of positioning error less than 10cm when the positioning device is positioned by method 4 is 92.0%.
  • the proportion of positioning error less than 10cm when the positioning device is positioned by method 1 is 56.3%, the proportion of positioning error is less than 10cm when positioning the positioning device by method 2 is 66.8%, method 4
  • the proportion of positioning error less than 10cm when the positioning device is positioned is 91.3%.
  • the proportion of positioning error less than 10cm when the positioning device is positioned by method 1 is 67.1%, the proportion of positioning error is less than 10cm when positioning the positioning device by method 2 is 85%, method 4
  • the proportion of positioning error less than 10cm when the positioning device is positioned is 94.2%. It can be seen that the positioning accuracy of method 4 is significantly better than that of method 1 and method 2.
  • the attribute information in the embodiment of the present application can also be used for environment perception of automatic driving.
  • features that are not easy to change in the environment can be extracted and used as attribute information to model the environment, or objects in the environment can be identified according to the process of acquiring semantic information in the embodiment of the present application to realize the movement of dynamic objects in the environment Estimation, thus providing conditions for path planning and obstacle avoidance of automatic driving.
  • An embodiment of the present application also provides a positioning device, which may be the positioning device in the above method embodiment, or a device including the above positioning device, or a component that can be used for the above positioning device.
  • the above-mentioned positioning device includes corresponding hardware structures and/or software modules for performing various functions.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software drives hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present application.
  • the functional modules of the positioning device may be divided according to the above method examples.
  • each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. It should be noted that the division of modules in the embodiment of the present application is schematic, and is only a logical function division, and there may be other division methods in actual implementation.
  • FIG. 13 shows a schematic structural diagram of a positioning device 130 .
  • the positioning device 130 includes an acquisition module 1301 and a processing module 1302 .
  • the positioning device 130 is used to realize the functions of the above positioning device.
  • the positioning device 130 is, for example, the terminal shown in the embodiment shown in FIG. 3 , the embodiment shown in FIG. 5 , the embodiment shown in FIG. 8 , the embodiment shown in FIG. 10 or the embodiment shown in FIG. 11 .
  • the acquiring module 1301 is configured to acquire a grid map corresponding to the first environment, the grid map includes a plurality of first grids, the first grid includes a plurality of points, position information of the plurality of points, and Attribute information, the attribute information of the first grid includes the semantic information of the first grid and the height information of the first grid, the semantic information of the first grid is used to indicate that the first grid represents each object in various objects The probability of , the height information of the first grid is used to indicate the height of the points included in the first grid.
  • the acquiring module 1301 is further configured to acquire a local grid corresponding to the second environment, the second environment is included in the first environment, the local grid includes at least one second grid, the second grid includes a plurality of points and the second grid
  • the attribute information of the second grid, the attribute information of the second grid includes the semantic information of the second grid and the height information of the second grid, and the semantic information of the second grid is used to indicate that the second grid represents each of the various objects
  • the probability of an object, and the height information of the second grid is used to indicate the height of the points included in the second grid.
  • the acquiring module 1301 is further configured to acquire a first matching probability, where the first matching probability is a matching probability between the local grid and the grid in the grid map when the local grid is located at different positions in the grid map.
  • the processing module 1302 is configured to determine the location of the device to be located according to the first matching probability, and the device to be located is located in the second environment.
  • the attribute information of the first grid further includes reflection intensity information of the first grid, and the reflection intensity information of the first grid is used to indicate the reflection intensity of points included in the first grid;
  • the attribute information of the second grid further includes reflection intensity information of the second grid, and the reflection intensity information of the second grid is used to indicate the reflection intensity of the points included in the second grid.
  • the processing module 1302 is further configured to predict the first position corresponding to the position of the device to be positioned at the second moment according to the pose information of the device to be positioned at the first moment and the motion parameters of the device to be positioned. area.
  • the acquiring module 1301 is specifically configured to acquire a first matching probability between the local grid and the grid in the grid map when the local grid is located in the grid in the first region.
  • the acquiring module 1301 is specifically configured to acquire each attribute information of the first grid in the first area according to the attribute information of the grid in the local grid and the attribute information of the grid in the first area. The first matching probability corresponding to the attribute information.
  • the acquisition module 1301 is specifically configured to acquire the first attribute information of the second grid in the local grid and the first attribute information of the second grid in the case where the local grid is located in the first region
  • the first deviation between the first attribute information of the first grid corresponding to the two grids, the first attribute information of the second grid is any one of the attribute information of the second grid, and the first attribute information of the first grid
  • the attributes of the first attribute information and the first attribute information of the second grid are the same;
  • the acquiring module 1301 is also specifically configured to acquire the first attribute corresponding to the first attribute information of the first grid in the first region according to the first deviation. match probability.
  • the processing module 1302 is specifically configured to determine the position information corresponding to each attribute information according to the first matching probability corresponding to each attribute information of the first grid in the first area; the processing module 1302 , which is also specifically used to determine the location of the device to be positioned according to the location information corresponding to each attribute information.
  • the processing module 1302 is further specifically configured to determine, among the location information corresponding to each type of attribute information, the location information corresponding to at least one attribute information conforming to preset rules; the processing module 1302 is also specifically configured to use If the quantity of at least one kind of attribute information is greater than 1, the position information corresponding to at least one kind of attribute information is weighted and summed to obtain the position of the device to be positioned; or, if the quantity of at least one kind of attribute information is equal to 1, then at least The location information corresponding to one kind of attribute information is determined as the location of the device to be positioned.
  • the acquisition module 1301 is also used to acquire the second matching probability.
  • the second matching probability is the local The matching probability of the grid and the grid in the grid map; the processing module 1302 is further configured to determine the orientation angle of the device to be located according to the second matching probability.
  • the processing module 1302 is further configured to predict the orientation angle of the device to be positioned at the second moment according to the orientation angle of the device to be positioned at the third moment and the motion parameters of the device to be positioned.
  • the acquiring module 1301 is specifically configured to acquire the second matching probabilities of multiple rotation angles corresponding to the local grid when the local grid is located at the position of the device to be positioned, among the multiple rotation angles , the included angle between each rotation angle and the predicted orientation angle of the device to be positioned at the second moment is less than or equal to the first angle.
  • the obtaining module 1301 is specifically configured to obtain the first rotation angle according to the attribute information of the grid in the local grid corresponding to the first rotation angle and the attribute information of the grid in the grid map
  • the second matching probability corresponding to each attribute information of , and the first rotation angle is any one of the multiple rotation angles.
  • the obtaining module 1301 is specifically configured to obtain the position of each second grid in the local grid when the local grid is located at the position of the device to be positioned and the local grid is rotated by the first rotation angle.
  • the first attribute information of the grid, the second deviation between the first attribute information of the first grid corresponding to the second grid, the first attribute information of the second grid is any of the attribute information of the second grid
  • a piece of attribute information, the first attribute information of the first grid is the same as the attribute of the first attribute information of the second grid
  • the acquisition module 1301 is also specifically configured to acquire the first attribute information of the first rotation angle according to the second deviation The corresponding second matching probability.
  • the processing module 1302 is specifically configured to determine the orientation angle information corresponding to each type of attribute information according to the second matching probability corresponding to each type of attribute information of each rotation angle; the processing module 1302 is also specifically configured to It is used to weight and sum the orientation angle information corresponding to each attribute information to obtain the orientation angle of the device to be positioned.
  • the acquiring module 1301 is specifically configured to acquire point cloud information corresponding to the first environment, and the point cloud information corresponding to the first environment is used to indicate the position of the midpoint of the point cloud corresponding to the first environment The information; acquisition module 1301 is specifically configured to acquire the grid map corresponding to the first environment according to the point cloud information corresponding to the first environment.
  • the information of the point cloud corresponding to the first environment is further used to indicate the reflection intensity of the points in the point cloud corresponding to the first environment.
  • the obtaining module 1301 is specifically configured to obtain point cloud information corresponding to the second environment, and the point cloud information corresponding to the second environment is used to indicate the position of the point cloud corresponding to the second environment
  • the information; obtaining module 1301 is also specifically configured to obtain the local grid corresponding to the second environment according to the point cloud information corresponding to the second environment.
  • the point cloud information corresponding to the second environment is further used to indicate the reflection intensity of the points in the point cloud corresponding to the second environment.
  • FIG. 3 When used to realize the functions of the above positioning device, other functions that can be realized by the positioning device 130 can be referred to the embodiment shown in FIG. 3 , the embodiment shown in FIG. 5 , the embodiment shown in FIG. 8 , and the embodiment shown in FIG. 10
  • FIG. 11 The illustrated embodiment or the related introduction of the embodiment shown in FIG. 11 will not be described in detail.
  • the positioning device 130 can take the form shown in FIG. 2 .
  • the processor 201 in FIG. 2 may invoke the computer-executed instructions stored in the memory 203, so that the positioning device 130 executes the methods described in the foregoing method embodiments.
  • the functions/implementation process of the acquisition module 1301 and the processing module 1302 in FIG. 13 may be implemented by the processor 201 in FIG. 2 invoking computer-executed instructions stored in the memory 203 .
  • one or more of the above modules or units may be implemented by software, hardware or a combination of both.
  • the software exists in the form of computer program instructions and is stored in the memory, and the processor can be used to execute the program instructions and realize the above method flow.
  • the processor can be built into a SoC (system on a chip) or ASIC, or it can be an independent semiconductor chip.
  • the core of the processor is used to execute software instructions for calculation or processing, and can further include necessary hardware accelerators, such as field programmable gate array (field programmable gate array, FPGA), PLD (programmable logic device) , or a logic circuit that implements a dedicated logic operation.
  • the hardware can be CPU, microprocessor, digital signal processing (digital signal processing, DSP) chip, microcontroller unit (microcontroller unit, MCU), artificial intelligence processor, ASIC, Any one or any combination of SoC, FPGA, PLD, dedicated digital circuit, hardware accelerator or non-integrated discrete device, which can run necessary software or not depend on software to execute the above method flow.
  • DSP digital signal processing
  • MCU microcontroller unit
  • ASIC artificial intelligence processor
  • an embodiment of the present application further provides a chip system, including: at least one processor and an interface, the at least one processor is coupled to the memory through the interface, and when the at least one processor executes the computer program or instruction in the memory When, the method in any one of the above method embodiments is executed.
  • the chip system further includes a memory.
  • the system-on-a-chip may consist of a chip, or may include a chip and other discrete devices, which is not specifically limited in this embodiment of the present application.
  • the embodiment of the present application further provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be completed by computer programs to instruct related hardware, and the program can be stored in the above computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments .
  • the computer-readable storage medium may be an internal storage unit of the positioning device in any of the foregoing embodiments, such as a hard disk or memory of the positioning device.
  • the above-mentioned computer-readable storage medium may also be an external storage device of the above-mentioned positioning device, such as a plug-in hard disk equipped on the above-mentioned positioning device, a smart memory card (smart media card, SMC), a secure digital (secure digital, SD) card, Flash card (flash card), etc. Further, the above-mentioned computer-readable storage medium may also include both an internal storage unit of the above-mentioned positioning device and an external storage device.
  • the above-mentioned computer-readable storage medium is used to store the above-mentioned computer program and other programs and data required by the above-mentioned positioning device.
  • the computer-readable storage medium described above can also be used to temporarily store data that has been output or will be output.
  • the embodiment of the present application further provides a computer program product. All or part of the processes in the above method embodiments can be completed by computer programs instructing related hardware, and the programs can be stored in the above computer program products. When the programs are executed, they can include the processes of the above method embodiments.
  • An optional embodiment of the present application further provides a computer instruction. All or part of the procedures in the above method embodiments may be completed by computer instructions to instruct related hardware (such as computers, processors, access network devices, mobility management network elements or session management network elements, etc.).
  • the program may be stored in the above-mentioned computer-readable storage medium or in the above-mentioned computer program product.
  • the disclosed devices and methods may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components can be Incorporation or may be integrated into another device, or some features may be omitted, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may be one physical unit or multiple physical units, that is, it may be located in one place, or may be distributed to multiple different places . Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Electromagnetism (AREA)
  • Navigation (AREA)

Abstract

本申请公开了定位方法及装置，涉及定位技术领域，可以提高定位精度和稳定性。该方法包括：获取第一环境对应的栅格地图和位于第一环境中的第二环境对应的局部栅格，获取局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的第一匹配概率，并根据该第一匹配概率确定待定位装置的位置。其中，栅格地图包括多个第一栅格，第一栅格包括多个点、多个点中每个点的位置信息和第一栅格的属性信息。局部栅格包括至少一个第二栅格，第二栅格包括多个点和第二栅格的属性信息。上述属性信息包括语义信息和高度信息。

Description

定位方法及装置 技术领域
本申请涉及定位技术领域，尤其涉及定位方法及装置。
背景技术
随着计算机技术和互联网技术的不断发展，智能驾驶得到了广泛关注。智能驾驶包括辅助驾驶和自动驾驶。它们的关键技术有定位，感知，规划决策，控制等等。其中，高精度定位可以为规划决策或控制提供可靠依据，因此，高精度定位在智能驾驶中发挥着不可或缺的作用。
目前，主要通过全球定位系统(global positioning system，GPS)来进行定位。然而，在GPS信号受遮挡或定位环境复杂的情况下，通过GPS定位会产生较大的定位误差，无法提供高精度和高稳定性的定位结果。
发明内容
本申请实施例提供定位方法及装置，可以提高定位精度和稳定性。
为达到上述目的，本申请的实施例采用如下技术方案：
第一方面，提供了一种定位方法，该方法包括：获取第一环境对应的栅格地图，获取第二环境对应的局部栅格，获取该局部栅格位于该栅格地图中的不同位置时，该局部栅格与该栅格地图中的栅格的第一匹配概率，根据该第一匹配概率确定待定位装置的位置。
其中，上述栅格地图包括多个第一栅格，任意一个第一栅格包括多个点、该多个点的位置信息和该第一栅格的属性信息。该第一栅格的属性信息包括该第一栅格的语义信息和该第一栅格的高度信息，该第一栅格的语义信息用于指示该第一栅格表示多种物体中的每种物体的概率，该第一栅格的高度信息用于指示该第一栅格包括的点的高度；该第二环境包括在该第一环境中，该局部栅格包括至少一个第二栅格，任意一个第二栅格包括多个点和该第二栅格的属性信息，该第二栅格的属性信息包括该第二栅格的语义信息和该第二栅格的高度信息，该第二栅格的语义信息用于指示该第二栅格表示多种物体中的每种物体的概率，该第二栅格的高度信息用于指示该第二栅格包括的点的高度。该待定位装置位于该第二环境中。
基于上述第一方面提供的定位方法，定位装置可以获取第一环境对应的栅格地图，获取第二环境信息对应的局部栅格，并结合第一栅格的多种属性信息和第二栅格的多种属性信息对待定位装置进行定位，提高了定位的准确性和稳定性。
在一种可能的实现方式中，该第一栅格的属性信息还包括该第一栅格的反射强度信息，该第一栅格的反射强度信息用于指示该第一栅格包括的点的反射强度；该第二栅格的属性信息还包括该第二栅格的反射强度信息，该第二栅格的反射强度信息用于指示该第二栅格包括的点的反射强度。基于上述方法，定位装置还可以结合第一栅格 的反射强度信息和第二栅格的反射强度信息进行定位，进一步提高定位的准确性和稳定性。
在一种可能的实现方式中，该方法还包括：根据该待定位装置在第一时刻的位姿信息和该待定位装置的运动参数，预测该待定位装置在第二时刻的位置对应的第一区域。基于上述方法，定位装置可以预测待定位装置在第二时刻的位置，以便根据预测的位置对待定位装置定位。
在一种可能的实现方式中，该获取第一匹配概率，包括：获取该局部栅格位于该第一区域中的栅格时，该局部栅格与该栅格地图中的栅格的第一匹配概率。基于上述方法，定位装置可以不需要将局部栅格中的栅格与栅格地图中的全部栅格进行匹配，而是将局部栅格中的栅格与第一区域中的栅格进行匹配，减少了计算开销。
在一种可能的实现方式中，获取该局部栅格位于该第一区域中的栅格时，该局部栅格与该栅格地图中的栅格的第一匹配概率，包括：根据该局部栅格中的栅格的属性信息和该第一区域中的栅格的属性信息，获取第一区域中第一栅格的每种属性信息对应的第一匹配概率。基于上述方法，定位装置可以基于第一栅格的属性信息和第二栅格的属性信息，确定第一区域中第一栅格的每种属性信息对应的第一匹配概率，进而可以根据第一匹配概率对待定位装置定位，使得定位装置在定位时不是仅依靠某一种属性来定位，提高了定位的准确性和稳定性。
在一种可能的实现方式中，根据该局部栅格中的栅格的属性信息和该第一区域中的栅格的属性信息，获取第一区域中第一栅格的第一属性信息对应的第一匹配概率，包括：获取该局部栅格位于该第一区域中的第一栅格的情况下，该局部栅格中的第二栅格的第一属性信息与该第二栅格对应的第一栅格的第一属性信息之间的第一偏差，该第二栅格的第一属性信息为该第二栅格的属性信息中的任意一个属性信息，该第一栅格的第一属性信息与该第二栅格的第一属性信息的属性相同；根据该第一偏差，获取该第一区域中的第一栅格的第一属性信息对应的第一匹配概率。基于上述方法，定位装置可以将局部栅格中的每个第二栅格和第一区域中的每个第一栅格进行匹配，以确定第一区域中的第一栅格的第一属性信息对应的第一匹配概率，使得对待定位装置定位的精度更高。
在一种可能的实现方式中，根据该第一匹配概率确定待定位装置的位置，包括：根据该第一区域中第一栅格的每种属性信息对应的第一匹配概率，确定每种属性信息对应的位置信息；根据该每种属性信息对应的位置信息，确定该待定位装置的位置。基于上述方法，定位装置可以根据第一区域中第一栅格的每种属性信息对应的第一匹配概率，获取每种属性信息对应的位置信息，并根据该位置信息，获取待定位装置的位置，从而对待定位装置进行定位。
在一种可能的实现方式中，根据该第一匹配概率确定该待定位装置的位置，还包括：在该每种属性信息对应的位置信息中确定符合预设规则的至少一种属性信息对应的位置信息；根据该每种属性信息对应的位置信息，确定该待定位装置的位置，包括：若该至少一种属性信息的数量大于1，则对该至少一种属性信息对应的位置信息加权求和，得到该待定位装置的位置；或者，若该至少一种属性信息的数量等于1，则将该至少一种属性信息对应的位置信息确定为该待定位装置的位置。基于上述方法，定 位装置在确定了每种属性信息对应的位置信息后，可以先对这些位置信息进行筛选，剔除不准确的位置信息，再根据筛选后的位置信息确定待定位装置的位置，以使得定位装置对待定位装置的定位更为准确。
在一种可能的实现方式中，该方法还包括：获取第二匹配概率，该第二匹配概率为在该局部栅格位于该待定位装置的位置，并且该局部栅格旋转不同角度的情况下，该局部栅格与该栅格地图中的栅格的匹配概率；根据该第二匹配概率确定该待定位装置的朝向角。基于上述方法，定位装置可以确定待定位装置的朝向角，即待定位装置的运动方向。
在一种可能的实现方式中，该方法还包括：根据该待定位装置在第三时刻的朝向角和该待定位装置的运动参数，预测该待定位装置在第二时刻的朝向角。基于上述方法，定位装置可以预测待定位装置在第二时刻的朝向角，以便根据预测的朝向角确定对待定位装置的朝向角。
在一种可能的实现方式中，获取该第二匹配概率，包括：获取该局部栅格位于该待定位装置的位置的情况下，该局部栅格对应的多个旋转角的第二匹配概率，该多个旋转角中，每个旋转角与该预测该待定位装置在第二时刻的朝向角之间的夹角小于或等于第一角度。基于上述方法，定位装置不需要将局部栅格旋转一圈来获取多个旋转角，而是将局部栅格在第二时刻的朝向角附近旋转来获取多个旋转角。既可以降低定位装置的计算开销，又可以提高确定的待定位装置的朝向角的精度。
在一种可能的实现方式中，获取该局部栅格位于该待定位装置的位置的情况下，第一旋转角的第二匹配概率，包括：根据该第一旋转角对应的局部栅格中的栅格的属性信息和该栅格地图中的栅格的属性信息，获取该第一旋转角的每种属性信息对应的第二匹配概率，该第一旋转角为该多个旋转角中的任意一个旋转角。基于上述方法，定位装置可以基于第一栅格的属性信息和第二栅格的属性信息，确定第一旋转角的每种属性信息对应的第二匹配概率，进而可以根据第二匹配概率确定待定位装置的朝向角，使得定位装置在确定待定位装置的朝向角时不是仅依靠某一种属性，提高了定位装置确定的朝向角的准确性。
在一种可能的实现方式中，根据该第一旋转角对应的局部栅格中的栅格的属性信息和该栅格地图中的栅格的属性信息，获取该第一旋转角的第一属性信息对应的第二匹配概率，包括：获取该局部栅格位于该待定位装置的位置，并且该局部栅格旋转第一旋转角的情况下，该局部栅格中的每个第二栅格的第一属性信息，与该第二栅格对应的第一栅格的第一属性信息之间的第二偏差，该第二栅格的第一属性信息为该第二栅格的属性信息中的任意一个属性信息，该第一栅格的第一属性信息与该第二栅格的第一属性信息的属性相同；根据该第二偏差，获取该第一旋转角的第一属性信息对应的第二匹配概率。基于上述方法，定位装置可以将局部栅格中的每个第二栅格和第一旋转角下，第二栅格对应的第一栅格进行匹配，以确定第一旋转角的第一属性信息对应的第二匹配概率，使得定位装置确定的朝向角更为准确。
在一种可能的实现方式中，根据该第二匹配概率确定待定位装置的朝向角，包括：根据每个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息；对该每种属性信息对应的朝向角信息加权求和，得到该待定位装置的朝向角。 基于上述方法，定位装置可以根据每个旋转角的每种属性信息对应的第二匹配概率，获取每种属性信息对应的朝向角信息，并根据该朝向角信息，获取待定位装置的朝向角，从而实现定位装置确定待定位装置的运动方向。
在一种可能的实现方式中，获取第一环境对应的栅格地图，包括：获取该第一环境对应的点云的信息，该第一环境对应的点云的信息用于指示该第一环境对应的点云中点的位置信息；根据该第一环境对应的点云的信息，获取该第一环境对应的栅格地图。基于上述方法，定位装置可以获取到第一环境对应的栅格地图。
在一种可能的实现方式中，第一环境对应的点云的信息还用于指示该第一环境对应的点云中点的反射强度。基于上述方法，定位装置可以根据第一环境对应的点云信息获取第一栅格的反射强度信息。
在一种可能的实现方式中，获取第二环境对应的局部栅格，包括：获取该第二环境对应的点云的信息，该第二环境对应的点云的信息用于指示该第二环境对应的点云中点的位置信息；根据该第二环境对应的点云的信息，获取该第二环境对应的局部栅格。基于上述方法，定位装置可以获取到第二环境对应的局部栅格。
在一种可能的实现方式中，该第二环境对应的点云的信息还用于指示该第二环境对应的点云中点的反射强度。定位装置可以根据第二环境对应的点云信息获取第二栅格的反射强度信息。
第二方面，提供了一种定位装置用于实现上述方法。该定位装置可以为上述第一方面中的定位装置，或者包含上述定位装置的装置。该定位装置包括实现上述方法相应的模块、单元、或手段(means)，该模块、单元、或means可以通过硬件实现，软件实现，或者通过硬件执行相应的软件实现。该硬件或软件包括一个或多个与上述功能相对应的模块或单元。
第三方面，提供了一种定位装置，包括：处理器；该处理器用于与存储器耦合，并读取存储器中的指令之后，根据该指令执行如上述第一方面所述的方法。该定位装置可以为上述第一方面中的定位装置，或者包含上述定位装置的装置。
结合上述第三方面，在一种可能的实现方式中，该定位装置还包括存储器，该存储器，用于保存必要的程序指令和数据。
结合上述第三方面，在一种可能的实现方式中，该定位装置为芯片或芯片系统。可选的，该定位装置是芯片系统时，可以由芯片构成，也可以包含芯片和其他分立器件。
第四方面，提供了一种定位装置，包括：处理器和接口电路；接口电路，用于接收计算机程序或指令并传输至处理器；处理器用于执行所述计算机程序或指令，以使该定位装置执执行如上述第一方面所述的方法。
结合上述第四方面，在一种可能的实现方式中，该定位装置为芯片或芯片系统。可选的，该定位装置是芯片系统时，可以由芯片构成，也可以包含芯片和其他分立器件。
第五方面，提供了一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机可以执行上述第一方面所述的方法。
第六方面，提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使 得计算机可以执行上述第一方面所述的方法。
其中，第二方面至第六方面中任一种可能的实现方式所带来的技术效果可参见上述第一方面或任一方面中不同可能的实现方式所带来的技术效果，此处不再赘述。
附图说明
图1A为本申请实施例提供的定位系统架构示意图一；
图1B为本申请实施例提供的定位系统架构示意图二；
图2为本申请实施例提供的定位装置的硬件结构示意图；
图3为本申请实施例提供的定位方法的流程示意图一；
图4为本申请实施例提供的栅格地图和局部栅格的示意图一；
图5为本申请实施例提供的定位方法的流程示意图二；
图6为本申请实施例提供的待定位装置的角速度、待定位装置的线速度和待定位装置车轮的转角的示意图；
图7为本申请实施例提供的栅格地图和局部栅格的示意图二；
图8为本申请实施例提供的定位方法的流程示意图三；
图9为本申请实施例提供的栅格地图和局部栅格的示意图三；
图10为本申请实施例提供的定位方法的流程示意图四；
图11为本申请实施例提供的定位方法的流程示意图五；
图12为本申请实施例提供的待定位装置的位置的示意图；
图13为本申请实施例提供的定位装置的结构示意图。
具体实施方式
首先，为了便于理解本申请实施例提供的技术方案，对本申请实施例涉及的技术术语进行解释说明。
1、同步建图与定位(simultaneous localization and mapping，SLAM)
SLAM可以指定位与地图构建，或并发建图与定位。具体的，可以将地图获取装置(如机器人)放入未知环境中的未知位置，让地图获取装置根据周围的环境信息，以及该地图获取装置在各个时刻的位置信息，一边移动一边逐步描绘出此环境的地图。
2、迭代最近邻点(iterative closest point，ICP)
ICP算法可以按照一定的约束条件，在源点云集合中确定与目标点云集合中的点集Pi距离最近的点集Qi，并计算旋转矩阵R和平移矩阵T，使得误差函数最小；对Pi使用R和T进行旋转和平移变换得到新的点P' i；计算P' i与Qi的平均距离d，若d小于某一个阈值，或者迭代计算旋转矩阵和平移矩阵的次数大于最大迭代次数，则停止计算，否则重新在源点云集合中确定与目标点云集合中的点集Pi距离最近的点集Q' i，并继续迭代计算旋转矩阵和平移矩阵，直至得到的平均距离小于某一个阈值，或者迭代计算旋转矩阵和平移矩阵的次数大于最大迭代次数。
3、激光雷达里程计和建图
激光雷达里程计和建图是一种利用激光雷达实现SLAM的算法。通过激光雷达点云的匹配计算机器人相对于初始时刻的位置，称为激光雷达里程计，根据每个位置把点云结合到一起构成地图，称为建图。
4、点云库(point cloud library，PCL)
PCL是在吸收了前人点云相关研究基础上建立起来的大型跨平台开源C++编程库。它实现了大量点云相关的通用算法和高效数据结构，涉及到点云获取、滤波、分割、配准、检索、特征提取、识别、追踪、曲面重建、可视化等。支持多种操作系统平台，可在Windows、Linux、Android、Mac OS X、部分嵌入式实时系统上运行。
5、PointNet
PointNet是一种新型的处理点云数据的深度学习模型。它能够用于点云数据的多种认知任务，如分类、语义分割和目标识别。
如背景技术中所述，在GPS信号受遮挡或定位环境复杂的情况下，通过GPS定位会产生较大的定位误差，无法提供高精度和高稳定性的定位结果。为了解决这个问题，可以通过如下方法进行定位，以提高定位的准确性和稳定性。
方法1：通过ICP算法定位
在方法1中，可以使用ICP算法将激光雷达实时获取的点云(可以相当于上述Pi)与预先建立的点云地图中的点云(可以相当于上述与目标点云集合中的点集Pi距离最近的点集)进行匹配，以确定实时位置。
方法2：基于特征信息和栅格地图的定位方法
在方法2中，可以根据激光雷达实时获取的点云信息(如点云的位置信息、点云的特征信息等)得到局部栅格，将局部栅格中每个栅格的特征信息与预先存储的栅格地图中的点云的特征信息进行匹配，从而完成定位。其中，点云的特征信息可以包括高度信息。预先存储的栅格地图是根据激光雷达在栅格地图对应的环境中获取的点云信息得到的。
方法3：基于种类信息和栅格地图的定位方法
在方法3中，可以根据激光雷达实时获取的点云的位置信息得到局部栅格，通过视觉传感器获取的上述点云对应的图像信息，并识别图像信息中的物体，得到物体对应的种类信息，将种类信息保存到局部栅格中；再将局部删格中每个栅格的种类信息与预先存储的栅格地图中的每个栅格的种类信息进行匹配，从而完成定位。其中，预先存储的栅格地图是根据激光雷达在栅格地图对应的环境中获取的点云信息，以及视觉传感器在栅格地图对应的环境中获取的图像信息得到的。
方法4：基于属性信息和栅格地图的定位方法
在方法4中，可以获取第一环境对应的栅格地图，获取第二环境信息对应的局部栅格。其中，栅格地图包括多个第一栅格，第一栅格包括多个点、多个点中每个点的位置信息和第一栅格的属性信息，第一栅格的属性信息包括第一栅格的语义信息和第一栅格的高度信息。第二环境包括在第一环境中，局部栅格包括至少一个第二栅格，第二栅格包括多个点和第二栅格的属性信息。第二栅格的属性信息包括第二栅格的语义信息和第二栅格的高度信息。后续，可以获取局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的第一匹配概率，根据第一匹配概率确定待定位装置的位置。方法4将在下述图3所示的实施例中进行阐述，此处不做赘述。
可以理解的，上述方法1虽然实施起来较为简单，但是方法1仅仅考虑了点和点之间的距离最小，算法容易收敛到错误的位置，即定位不准确。上述方法2结合了点 云的特征信息来定位，相比于方法1，提高了定位精度。但是，若该特征信息发生改变，会导致方法2的定位不准确，稳定性差。上述方法3结合了物体的种类信息来定位，相比于方法1，提高了定位精度。但是，方法3通常依赖识别图像信息或点云信息中的物体的准确性，若识别物体的准确度低，则定位不准确。上述方法4可以结合栅格的属性信息，如栅格的高度信息和栅格的语义信息进行定位，相比于方法1-方法3，通过方法4进行定位的准确性和稳定性都更高。
可以理解的，本申请实施例提供定位方法及装置，能够在一定区域内进行定位。例如，该定位方法及装置可以在居民小区、工业园区、公园等环境中定位。通过本申请实施例提供的定位方法及装置，可以提高定位精度和稳定性。
下面对本申请实施例可应用的系统架构进行说明。
在一种可能的实现方式中，本申请实施例可应用的系统架构包括定位装置。可选的，该系统架构还包括感知装置。感知装置可以通过有线或无线的方式与定位装置通信。示例性的，该系统架构可以如图1A所示。图1A所示的系统架构10包括定位装置101。可选的，该系统架构10还包括感知装置102。
其中，上述定位装置可以用于获取第一环境对应的栅格地图，获取第二环境对应的局部栅格，获取局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的第一匹配概率，并根据第一匹配概率确定待定位装置的位置。其中，栅格地图包括多个第一栅格。任意一个第一栅格包括多个点、多个点中每个点的位置信息和第一栅格的属性信息。第一栅格的属性信息包括第一栅格的语义信息和第一栅格的高度信息。第二环境包括在第一环境中，局部栅格包括至少一个第二栅格。任意一个第二栅格包括多个点和第二栅格的属性信息。第二栅格的属性信息包括第二栅格的语义信息和第二栅格的高度信息。
作为一种示例，上述定位装置可以为车载设备或计算设备。例如，上述定位装置为电脑、电脑中的运算单元、嵌入式设备、x86系统的工控机或无人驾驶中的无线终端等。上述定位装置可以是作为一个或多个部件或者单元而内置于车辆的车载模块、车载模组、车载部件、车载芯片或者车载单元，车辆通过内置的所述车载模块、车载模组、车载部件、车载芯片或者车载单元可以实施本申请的方法。
上述感知装置可以用于获取第一环境对应的点云的信息，并将该点云的信息发送给上述定位装置。其中，第一环境对应的点云的信息可以用于指示第一环境对应的点云中点的位置信息。可选的，第一环境对应的点云的信息还可以用于指示第一环境对应的点云中点的反射强度。上述感知装置还可以用于获取第二环境对应的点云的信息，并将该点云的信息发送给上述定位装置。其中，第二环境对应的点云的信息用于指示第二环境对应的点云中点的位置信息。可选的，第二环境对应的点云的信息还可以用于指示第二环境对应的点云中点的反射强度。作为一种示例，上述感知装置可以为激光雷达，或其他可以获取某一环境对应的点云的信息的装置。
在另一种可能的实现方式中，本申请实施例可应用的系统架构包括定位装置。该定位装置中可以部署定位模块。可选的，该定位装置中还部署感知模块。示例性的，该系统架构可以如图1B所示。图1B所示的系统架构11包括定位装置111。该定位装置111中部署定位模块1111。可选的，该定位装置111中还部署感知模块1112。
其中，上述定位模块可以用于获取第一环境对应的栅格地图，获取第二环境对应的局部栅格，获取局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的第一匹配概率，并根据第一匹配概率确定待定位装置的位置。上述感知模块可以用于获取第一环境对应的点云的信息，并将该点云的信息发送给上述定位模块。上述感知模块还可以用于获取第二环境对应的点云的信息，并将该点云的信息发送给上述定位模块。
作为一种示例，上述定位装置可以为具备感知功能和计算功能的设备。例如，上述定位装置为智能车，如无人驾驶车、自动外卖车、自动快递车等等。其中，定位模块为定位装置中具备计算功能的模块，感知模块为定位装置中具备感知功能的模块。以定位装置为智能车为例，定位模块可以为智能车中的车载电脑、车载电脑中的运算单元、智能车中的嵌入式设备、智能车中的工控机或智能车中的无线终端等；感知模块可以为智能车中的激光雷达，或智能车中的其他可以获取某一环境对应的点云的信息的装置。
图1A和图1B仅为示意图，并不构成对本申请提供的技术方案的适用场景的限定。本领域的技术人员应当明白，在具体实现过程中，上述系统架构还可以包括其他设备，同时也可根据具体需要来确定定位装置、感知装置、定位模块或感知模块的数量，不予限制。
可选的，本申请实施例图1A或图1B中的装置(例如定位装置)可以是一个通用设备或者是一个专用设备，本申请实施例对此不作具体限定。
可选的，本申请实施例图1A或图1B中的装置(例如定位装置)的相关功能可以由一个设备实现，也可以由多个设备共同实现，还可以是由一个设备内的一个或多个功能模块实现，本申请实施例对此不作具体限定。可以理解的是，上述功能既可以是硬件设备中的网络元件，也可以是在专用硬件上运行的软件功能，或者硬件与软件的结合，或者平台(例如，云平台)上实例化的虚拟化功能。
在具体实现时，本申请实施例图1A或图1B中的装置(例如定位装置)都可以采用图2所示的组成结构，或者包括图2所示的部件。图2所示为可适用于本申请实施例的定位装置的硬件结构示意图。该定位装置20包括至少一个处理器201，用于实现本申请实施例提供的方法。该定位装置20还可以包括通信线路202、存储器203和至少一个通信接口204。
处理器201可以是一个通用中央处理器(central processing unit，CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制本申请方案程序执行的集成电路。
通信线路202可包括一通路，在上述组件之间传送信息，例如总线。
通信接口204，用于与其他设备或通信网络通信。通信接口204可以是任何收发器一类的装置，如可以是以太网接口、无线接入网(radio access network，RAN)接口、无线局域网(wireless local area networks，WLAN)接口、收发器、管脚、总线、或收发电路等。
存储器203可以是只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM) 或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器可以是独立存在，通过通信线路202与处理器201相耦合。存储器203也可以和处理器201集成在一起。本申请实施例提供的存储器通常可以具有非易失性。
其中，存储器203用于存储执行本申请实施例提供的方案所涉及的计算机执行指令，并由处理器201来控制执行。处理器201用于执行存储器203中存储的计算机执行指令，从而实现本申请实施例提供的方法。或者，可选的，本申请实施例中，也可以是处理器201执行本申请下述实施例提供的方法中的处理相关的功能，通信接口204负责与其他设备或通信网络通信，本申请实施例对此不作具体限定。
可选的，本申请实施例中的计算机执行指令也可以称之为应用程序代码，本申请实施例对此不作具体限定。
本申请实施例中的耦合是装置、单元或模块之间的间接耦合或通信连接，可以是电性，机械或其它的形式，用于装置、单元或模块之间的信息交互。
作为一种实施例，处理器201可以包括一个或多个CPU，例如图2中的CPU0和CPU1。
作为一种实施例，定位装置20可以包括多个处理器，例如图2中的处理器201和处理器207。这些处理器中的每一个可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。这里的处理器可以指一个或多个设备、电路、和/或用于处理数据(例如计算机程序指令)的处理核。
可以理解的，图2中示出的组成结构并不构成对该定位装置的限定，除图2所示部件之外，该定位装置可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
下面将结合附图，对本申请实施例提供的定位方法进行描述。下述实施例中的定位装置可以具备图2所示部件，不予赘述。
需要说明的是，本申请实施例提供的定位方法可以应用于多个领域，例如：无人驾驶领域、自动驾驶领域、辅助驾驶领域、智能驾驶领域、网联驾驶领域、智能网联驾驶领域、汽车共享领域等。
需要说明的是，在本申请实施例中，“/”可以表示前后关联的对象是一种“或”的关系，例如，A/B可以表示A或B；“和/或”可以用于描述关联对象存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况，其中A，B可以是单数或者复数。此外，类似于“A、B和C中的至少一项”或“A、B或C中的至少一项”的表述通常用于表示如下中任一项：单独存在A；单独存在B；单独存在C；同时存在A和B；同时存在A和C；同时存在B和C；同时存在A、B和C。以上是以A、B和C共三个元素进行举例来说明该项目的可选用条目，当表述中具有更多元素时，该表述的含义可以按照前述规则获得。
为了便于描述本申请实施例的技术方案，在本申请实施例中，可以采用“第一”、“第二”等字样对功能相同或相似的技术特征进行区分。该“第一”、“第二”等字样并不对数量和执行次序进行限定，并且“第一”、“第二”等字样也并不限定一定不同。在本申请实施例中，“示例性的”或者“例如”等词用于表示例子、例证或说明，被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念，便于理解。
需要说明的是，在本申请实施例中，对于一种技术特征，通过“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”等区分该种技术特征中的技术特征，该“第一”、“第二”、“第三”、“A”、“B”、“C”和“D”描述的技术特征间无先后顺序或者大小顺序。
可以理解的，本申请实施例中同一个步骤或者具有相同功能的步骤或者技术特征在不同实施例之间可以互相参考借鉴。
可以理解的，本申请实施例中，定位装置可以执行本申请实施例中的部分或全部步骤，这些步骤仅是示例，本申请实施例还可以执行其它步骤或者各种步骤的变形。此外，各个步骤可以按照本申请实施例呈现的不同的顺序来执行，并且有可能并非要执行本申请实施例中的全部步骤。
如图3所示，为本申请实施例提供的一种定位方法，该定位方法包括S301-S304。
S301：定位装置获取第一环境对应的栅格地图。
其中，定位装置可以为图1A中的定位装置101，或图1B中的定位装置111。
上述第一环境可以为待定位装置所处的环境。例如，第一环境为居民小区、工业园区或公园等。待定位装置为需要被定位的装置，待定位装置与定位装置可以相同也可以不同。若待定位装置与定位装置相同，则表示定位装置需要对自己定位。若待定位装置与定位装置不同，则表示定位装置需要对除自己之外的装置定位。
上述栅格地图为二维地图。该栅格地图包括多个第一栅格。第一栅格可以为正方形或长方形，或者，多个第一栅格中的一部分栅格为正方形的，另一部分栅格为长方形的，不予限制。第一栅格的大小可以是固定的，即定位装置中预先存储好的，也可以是用户根据需要设置的。本申请实施例中，栅格也可以有其他的名称，例如，栅格还可以称为网格、区域等，不予限制。
作为一种示例，栅格地图可以如图4所示。图4中的栅格地图401包括63个第一栅格。栅格地图401中的第一栅格为正方形。
一种可能的实现方式，多个第一栅格中的任意一个第一栅格包括多个点、该多个点中每个点的位置信息和第一栅格的属性信息。
其中，点的位置信息可以用于指示该点在栅格地图中的位置。例如，点的位置信息包括该点在栅格地图中的横坐标的值x和纵坐标的值y，即(x，y)。
一种可能的实现方式，第一栅格包括的每个点的位置信息可以通过第一栅格中的每个点分布在第一栅格的不同位置呈现在第一栅格中。例如，点1的位置信息为(x1，y1)，点2的位置信息为(x2，y2)，点1和点2处于第一栅格中的不同位置。
第一栅格的属性信息可以包括第一栅格的语义信息和第一栅格的高度信息。可选 的，第一栅格的属性信息还可以包括第一栅格的反射强度信息。下面对第一栅格的属性信息包括的每种属性信息进行介绍。
第一栅格的语义信息可以用于指示第一栅格表示多种物体中的每种物体的概率。其中，第一栅格表示多种物体中的每种物体的概率可以理解为第一栅格中的点所属的物体(或第一栅格中的点所对应的物体)为多种物体中的每种物体的概率。该多种物体可以包括树木、停的车辆、墙壁、地面、建筑物、草坪或其他物体中的至少两种。示例性的，第一栅格的语义信息可以用于指示第一栅格表示树木的概率、表示停的车辆的概率、表示墙壁的概率、表示地面的概率、表示草坪的概率和表示其他物体的概率，或者说，第一栅格的语义信息可以用于指示第一栅格中的点所属的物体为树木的概率、为停的车辆的概率、为墙壁的概率、为地面的概率、为草坪的概率和为其他物体的概率。
作为一种示例，第一栅格的语义信息包括第一栅格表示多种物体中的每种物体的概率。以多种物体包括树木、停的车辆、建筑物和其他物体为例，若第一栅格表示树木的概率为0.95，表示停的车辆的概率为0.02，表示建筑物的概率为0.01，表示其他物体的概率为0.02，则第一栅格的语义信息包括{0.95,0.02,0.01,0.02}。
作为另一种示例，第一栅格的语义信息包括多种物体中每种物体的标签信息，以及第一栅格表示多种物体中的每种物体的概率。以多种物体包括树木、停的车辆、建筑物和其他物体为例，若第一栅格表示树木的概率为0.95，表示停的车辆的概率为0.02，表示建筑物的概率为0.01，表示其他物体的概率为0.02，则第一栅格的语义信息包括{树木：0.95；停的车辆：0.02；建筑物：0.01；其他物体：0.02}，或者第一栅格的语义信息包括{树木，0.95}；{停的车辆：0.02}；{建筑物：0.01}；{其他物体：0.02}。
一种可能的实现方式，第一栅格的语义信息可以通过第一栅格中的点的颜色属性参数呈现在第一栅格中。其中，颜色属性参数包括灰度、饱和度、明度或色相中的至少一项。以颜色属性参数包括灰度为例，多种物体中的每种物体可以对应一个灰度值，第一栅格中的点的灰度值为第一物体对应的灰度值，第一物体为第一栅格表示多种物体中的每种物体的概率中、概率最大的物体。
作为一种示例，以多种物体包括树木、停的车辆、建筑物和其他物体，树木对应的灰度值1，停的车辆对应灰度值2，建筑物对应灰度值3，其他物体对应灰度值4为例，若第一栅格表示树木的概率为0.95，表示停的车辆的概率为0.02，表示建筑物的概率为0.01，表示其他物体的概率为0.02，则第一栅格表示多种物体中的每种物体的概率中、概率最大的物体为树木，所以第一栅格中点的灰度值为树木对应的灰度值1。
第一栅格的高度信息可以用于指示第一栅格包括的点的高度。示例性的，第一栅格的高度信息包括第一栅格包括的所有点的高度的平均值。其中，第一栅格包括的所有点的高度的平均值
Figure PCTCN2021124206-appb-000001
z a为第一栅格包括的所有点中第a个点的高度，N为第一栅格包括的所有点的数量。
一种可能的实现方式，第一栅格的高度信息可以通过第一栅格中的点的颜色属性参数呈现在第一栅格中。以颜色属性参数包括明度为例，可以将高度划分为多个高度区间，任意两个高度区间包括的高度值不同，每个高度区间对应一个明度，那么，第 一栅格中的点对应的明度为该点的高度所属的高度区间对应的明度。
第一栅格的反射强度信息可以用于指示第一栅格包括的点的反射强度，例如，点所属的物体表面的反射率。示例性的，第一栅格的反射强度信息包括第一栅格包括的所有点的反射强度的平均值。其中，第一栅格包括的所有点的反射强度的平均值
Figure PCTCN2021124206-appb-000002
A a为第一栅格包括的所有点中第a个点的反射强度。
一种可能的实现方式，第一栅格的反射强度信息可以通过第一栅格中的点的颜色属性参数呈现在第一栅格中。以颜色属性参数包括色相为例，可以将反射强度划分为多个反射强度区间，任意两个反射强度区间包括的反射强度值不同，每个反射强度区间对应一个色相，那么，第一栅格中的点对应的色相为该点的反射强度所属的反射强度区间对应的色相。
一种可能的实现方式，定位装置可以至少通过如下两种示例性的方式获取第一环境对应的栅格地图。
方式1：定位装置获取第一环境对应的点云的信息，根据第一环境对应的点云的信息，获取第一环境对应的栅格地图。
其中，第一环境对应的点云的信息可以用于指示第一环境对应的点云中点的位置信息。可选的，第一环境对应的点云的信息还可以用于指示第一环境对应的点云中点的反射强度。方式1的具体过程将在后续图10所示的方法中进行阐述，在此不做赘述。
方式2：定位装置从本地获取第一环境对应的栅格地图。
作为一种示例，第一环境对应的栅格地图存储在本地。例如，定位装置从其他装置获取到栅格地图后，存储在本地。又例如，定位装置根据上述方法1获取到栅格地图后，存储在本地。后续，在需要使用该栅格地图时，定位装置从本地获取第一环境对应的栅格地图。
S302：定位装置获取第二环境对应的局部栅格。
其中，第二环境包括在第一环境中，也就是说，第二环境为第一环境中的一部分。第二环境为待定位装置当前所处的位置的周围环境，也就是说，待定位装置处于第二环境中。
示例性的，以第一环境为居民小区为例，第二环境可以为居民小区中的2号楼北侧；以第一环境为工业园区为例，第二环境可以为工业园区中的办公楼门前；以第一环境为公园为，第二环境可以为公园中的假山下面。
上述局部栅格可以包括至少一个第二栅格。第二栅格可以为正方形或长方形，或者，至少一个第二栅格中的一部分栅格为正方形的，另一部分栅格为长方形的，不予限制。第二栅格的大小可以是固定的，即定位装置中预先存储好的，也可以是用户根据需要设置的。
作为一种示例，局部栅格可以如图4所示。图4中的局部栅格402包括9个第二栅格。局部栅格402中的第二栅格为正方形。
一种可能的实现方式，至少一个第二栅格中的任意一个第二栅格包括多个点和第二栅格的属性信息。
其中，第二栅格的属性信息可以包括第二栅格的语义信息和第二栅格的高度信息。 可选的，第二栅格的属性信息还可以包括第二栅格的反射强度信息。
其中，第二栅格的语义信息可以用于指示第二栅格表示多种物体中的每种物体的概率。第二栅格表示多种物体中的每种物体的概率可以理解为第二栅格中的点所属的物体(或第二栅格中的点所对应的物体)为多种物体中的每种物体的概率。第二栅格的高度信息可以用于指示第二栅格包括的点的高度。第二栅格的反射强度信息可以用于指示第二栅格包括的点的反射强度。可以理解的，第二栅格的属性信息包括的内容与第一栅格的属性信息包括的内容类似，因此，第二栅格的属性信息的介绍可以参考上述S301中对第一栅格的属性信息的描述，在此不做赘述。
一种可能的实现方式，定位装置获取第二环境对应的点云的信息，根据第二环境对应的点云的信息，获取第二环境对应的局部栅格。其中，第二环境对应的点云的信息可以用于指示第二环境对应的点云中点的位置信息。可选的，第二环境对应的点云的信息还可以用于指示第二环境对应的点云中点的反射强度。上述具体过程将在后续图11所示的方法中进行阐述，在此不做赘述。
S303：定位装置获取第一匹配概率。
其中，第一匹配概率为局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的匹配概率。
可以理解的，上述第一栅格或第二栅格对应有多种属性信息。定位装置可以基于每种属性信息确定每种属性信息对应的第一匹配概率。
一种可能的实现方式，定位装置根据局部栅格中的栅格的属性信息和栅格地图中的栅格的属性信息，获取不同位置的每种属性信息对应的第一匹配概率。具体的，定位装置获取局部栅格位于栅格地图中的第一栅格的情况下，局部栅格中的第二栅格的第一属性信息与第二栅格对应的第一栅格的第一属性信息之间的第一偏差，根据第一偏差，获取栅格地图中的第一栅格的第一属性信息对应的第一匹配概率。
其中，第二栅格的第一属性信息为第二栅格的属性信息中的任意一个属性信息。例如，第二栅格的第一属性信息为第二栅格的语义信息，或第二栅格的高度信息或第二栅格的反射强度信息。
第一栅格的第一属性信息与第二栅格的第一属性信息的属性相同。例如，若第二栅格的第一属性信息为第二栅格的语义信息，则第一栅格的第一属性信息为第一栅格的语义信息；若第二栅格的第一属性信息为第二栅格的高度信息，则第一栅格的第一属性信息为第一栅格的高度信息；若第二栅格的第一属性信息为第二栅格的反射强度信息，则第一栅格的第一属性信息为第一栅格的反射强度信息。
可选的，定位装置还可以获取栅格地图中的第一栅格的每种属性信息对应的第一预测概率，以便定位装置根据栅格地图中的第一栅格的每种属性信息对应的第一匹配概率和第一预测概率，确定该第一栅格的每种属性信息对应的第一后验概率，进而根据该第一栅格的每种属性信息对应的第一后验概率确定待定位装置的位置。
其中，栅格地图中的任意一个第一栅格的第一属性信息对应的第一预测概率为根据第一属性信息预测的局部栅格当前位于该第一栅格的概率。栅格地图中的任意一个第一栅格的第一属性信息对应的第一后验概率为该第一栅格的第一属性信息对应的第一预测概率与该第一栅格的第一属性信息对应的第一匹配概率之积。
下面分别介绍第一属性信息为语义信息、高度信息和反射强度信息时，定位装置获取局部栅格位于栅格地图中的第一栅格时，该第一栅格的第一属性信息对应的第一匹配概率和第一预测概率的过程。
一种可能的实现方式，定位装置获取局部栅格位于栅格地图中的每个第一栅格时，每个第一栅格的第一属性信息对应的第一匹配概率和第一预测概率。具体的，可以参考下述示例1-示例3中的描述。
示例1：第一属性信息为语义信息时，定位装置获取局部栅格位于栅格地图中的每个第一栅格时，每个第一栅格的第一属性信息对应的第一匹配概率和第一预测概率的过程。
以图4所示的栅格地图401和局部栅格402为例，定位装置可以使得局部栅格402的中心逐一位于栅格地图401中的63个第一栅格的中心，以获取局部栅格402中的9个第二栅格的语义信息分别与9个第二栅格对应的9个第一栅格的语义信息之间的第一偏差。
若定位装置使得局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则局部栅格402相当于与栅格地图401中的阴影部分的栅格重叠。局部栅格402中第二栅格(1,1)对应的第一栅格为栅格地图401中的第一栅格(2,7)，局部栅格402中第二栅格(1,2)对应的第一栅格为栅格地图401中的第一栅格(2,8)，局部栅格402中第二栅格(1,3)对应的第一栅格为栅格地图401中的第一栅格(2,9)，局部栅格402中第二栅格(2,1)对应的第一栅格为栅格地图401中的第一栅格(3,7)，局部栅格402中第二栅格(2,2)对应的第一栅格为栅格地图401中的第一栅格(3,8)，以此类推。其中，第二栅格(a，b)表示局部栅格402中第a行，第b列的栅格，类似的，第一栅格(a，b)表示栅格地图401中第a行，第b列的栅格。
对于第二栅格(1,1)和第一栅格(2,7)，第二栅格(1,1)的语义信息和第一栅格(2,7)的语义信息的第一偏差满足公式
Figure PCTCN2021124206-appb-000003
其中，s 1为第二栅格(1,1)的语义信息和第一栅格(2,7)的语义信息的第一偏差，n为S301中的多种物体的数量，p b为第二栅格表示多种物体中的第二物体的概率，
Figure PCTCN2021124206-appb-000004
为第一栅格表示多种物体中的第二物体的概率。第二物体为多种物体中的任意一个物体。例如，若多种物体包括树木、建筑物和其他物体，第二栅格表示树木的概率为0.2，表示建筑物的概率为0.85，表示其他物体的概率为0.05，第一栅格表示树木的概率为0.95，表示建筑物的概率为0.01，表示其他物体的概率为0.04，则s 1＝(0.2-0.95)+(0.85-0.01)+(0.05-0.04)。采用与上述方法相同的方法，定位装置可以获取第二栅格(1,2)的语义信息和第一栅格(2,8)的语义信息的第一偏差s 2，获取第二栅格(1,3)的语义信息和第一栅格(2,9)的语义信息的第一偏差s 3，获取第二栅格(2,1)的语义信息和第一栅格(3,7)的语义信息的第一偏差s 4，…，获取第二栅格(3,3)的语义信息和第一栅格(4,9)的语义信息的第一偏差s 9。后续，定位装置可以获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，第二栅格的语义信息与第一栅格的语义信息的第一总偏差，即第一总偏差
Figure PCTCN2021124206-appb-000005
后续，定位装置可以根据第一总偏差获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，黑色的第一栅格的语义信息对应的第一匹配概率。该第一匹配概率p(z|x i,y i,m)可以满足公式：p(z|x i,y i,m)＝μe -S。其中，x i为 栅格地图401中黑色的第一栅格中心位置的横坐标，y i为栅格地图401中黑色的第一栅格中心位置的纵坐标。μ为常数。
可以理解的，定位装置可以采用与上述方法相同的方法，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，第二栅格的语义信息与第一栅格的语义信息的第一总偏差，根据每个第一总偏差，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，每个第一栅格的语义信息对应的第一匹配概率。
一种可能的实现方式，若定位装置是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的语义信息对应的第一预测概率为预设的初始值，如1与栅格地图中的第一栅格的数量之比，即任意一个第一栅格的语义信息对应的第一预测概率满足公式
Figure PCTCN2021124206-appb-000006
其中，
Figure PCTCN2021124206-appb-000007
为任意一个第一栅格的语义信息对应的第一预测概率，x i为该第一栅格的中心的横坐标，y i为该第一栅格的中心的纵坐标，num为栅格地图中的第一栅格的数量。
另一种可能的实现方式，若定位装置不是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的语义信息对应的第一预测概率满足公式：
Figure PCTCN2021124206-appb-000008
其中，η为常数。p(x 0,y 0|z,m)为最近一次获取的、待定位装置在栅格地图的位置为(x 0,y 0)时的语义信息对应的第一后验概率，(x 0,y 0)为定位装置根据待定位装置在上一时刻的位置信息和待定位装置的运动参数，预测的待定位装置在当前时刻在栅格地图中所处的位置。定位装置根据待定位装置在上一时刻的位置信息和待定位装置的运动参数，预测的待定位装置在当前时刻在栅格地图中所处的位置的过程可以参考后续图5所示的方法中所述，在此不做赘述。σ为方差。
本申请实施例中，局部栅格的中心还可以替换为其他位置，如局部栅格的左上角(如局部栅格402中第二栅格(1,1)的左上角)，局部栅格的右上角(如局部栅格402中第二栅格(1,3)的右上角)，局部栅格的左下角(如局部栅格402中第二栅格(3,1)的左下角)，局部栅格的右下角(如局部栅格402中第二栅格(3,3)的右上角)等。类似的，第一栅格的中心也可以替换为其他位置。例如，若局部栅格的中心替换为局部栅格的左上角，则第一栅格的中心可以替换为第一栅格的左上角(如栅格地图401中的第一栅格(2,7)的左上角)；若局部栅格的中心替换为局部栅格的右上角，则第一栅格的中心可以替换为第一栅格的右上角(如栅格地图401中的第一栅格(2,7)的右上角)；若局部栅格的中心替换为局部栅格的左下角，则第一栅格的中心可以替换为第一栅格的左下角(如栅格地图401中的第一栅格(2,7)的左下角)；若局部栅格的中心替换为局部栅格的右下角，则第一栅格的中心可以替换为第一栅格的右下角(如栅格地图401中的第一栅格(2,7)的右下角)。
示例2：第一属性信息为高度信息时，定位装置获取局部栅格位于栅格地图中的每个第一栅格时，每个第一栅格的第一属性信息对应的第一匹配概率和第一预测概率的过程。
以图4所示的栅格地图401和局部栅格402为例，与定位装置获取局部栅格402中的9个第二栅格的语义信息分别与9个第二栅格对应的9个第一栅格的语义信息之 间的第一偏差类似，定位装置获取可以在使得局部栅格402的中心逐一位于栅格地图401中的63个第一栅格的中心时，获取局部栅格402中的9个第二栅格的高度信息分别与9个第二栅格对应的9个第一栅格的高度信息之间的第一偏差。
若定位装置使得局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则局部栅格402相当于与栅格地图401中的阴影部分的栅格重叠。对于第二栅格(1,1)和第一栅格(2,7)，第二栅格(1,1)的高度信息和第一栅格(2,7)的高度信息的第一偏差满足公式：
Figure PCTCN2021124206-appb-000009
其中，
Figure PCTCN2021124206-appb-000010
为第二栅格(1,1)的高度信息和第一栅格(2,7)的高度信息的第一偏差，h 1为第二栅格(1,1)包括的所有点的高度的平均值，
Figure PCTCN2021124206-appb-000011
为第一栅格(2,7)包括的所有点的高度的平均值。采用与上述方法相同的方法，定位装置可以获取第二栅格(1,2)的高度信息和第一栅格(2,8)的高度信息的第一偏差
Figure PCTCN2021124206-appb-000012
获取第二栅格(1,3)的高度信息和第一栅格(2,9)的高度信息的第一偏差
Figure PCTCN2021124206-appb-000013
获取第二栅格(2,1)的高度信息和第一栅格(3,7)的高度信息的第一偏差
Figure PCTCN2021124206-appb-000014
获取第二栅格(3,3)的高度信息和第一栅格(4,9)的高度信息的第一偏差
Figure PCTCN2021124206-appb-000015
后续，定位装置可以获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，第二栅格的高度信息与第一栅格的高度信息的第一总偏差，即第一总偏差
Figure PCTCN2021124206-appb-000016
后续，定位装置可以根据第一总偏差获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，黑色的第一栅格的高度信息对应的第一匹配概率。该第一匹配概率q(z|x i,y i,m)可以满足公式：q(z|x i,y i,m)＝μe -H。
可以理解的，定位装置可以采用与上述方法相同的方法，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，第二栅格的高度信息与第一栅格的高度信息的第一总偏差，根据每个第一总偏差，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，每个第一栅格的高度信息对应的第一匹配概率。
一种可能的实现方式，若定位装置是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的高度信息对应的第一预测概率为预设的初始值，如1与栅格地图中的第一栅格的数量之比，即任意一个第一栅格的高度信息对应的第一预测概率满足公式
Figure PCTCN2021124206-appb-000017
其中，
Figure PCTCN2021124206-appb-000018
为任意一个第一栅格的高度信息对应的第一预测概率，x i为该第一栅格的中心的横坐标，y i为该第一栅格的中心的纵坐标。
另一种可能的实现方式，若定位装置不是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的高度信息对应的第一预测概率满足公式：
Figure PCTCN2021124206-appb-000019
其中，η为常数。q(x 0,y 0|z,m)为最近一次获取的、待定位装置在栅格地图的位置为(x 0,y 0)时的高度信息对应的第一后验概率，(x 0,y 0)为定位装置根据待定位装置在上一时刻的位置信息和待定位装置的运动参数，预测的待定位装置在当前时刻在栅格地图中所处的位置。σ为方差。
示例3：第一属性信息为反射强度信息时，定位装置获取局部栅格位于栅格地图中的每个第一栅格时，每个第一栅格的第一属性信息对应的第一匹配概率和第一预测概率的过程。
可以理解的，第一属性信息为反射强度信息的情况下，定位装置获取每个第一栅 格的第一属性信息对应的第一匹配概率的过程，与第一属性信息为高度信息的情况下，定位装置获取每个第一栅格的第一属性信息对应的第一匹配概率的过程类似。
以图4所示的栅格地图401和局部栅格402为例，定位装置获取可以在使得局部栅格402的中心逐一位于栅格地图401中的63个第一栅格的中心时，获取局部栅格402中的9个第二栅格的反射强度信息分别与9个第二栅格对应的9个第一栅格的反射强度信息之间的第一偏差。
若定位装置使得局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则局部栅格402相当于与栅格地图401中的阴影部分的栅格重叠。对于第二栅格(1,1)和第一栅格(2,7)，第二栅格(1,1)的反射强度信息和第一栅格(2,7)的反射强度信息的第一偏差可以满足公式：
Figure PCTCN2021124206-appb-000020
其中，
Figure PCTCN2021124206-appb-000021
为第二栅格(1,1)的反射强度信息和第一栅格(2,7)的反射强度信息的第一偏差，a 1为第二栅格(1,1)包括的所有点的反射强度的平均值，
Figure PCTCN2021124206-appb-000022
为第一栅格(2,7)包括的所有点的反射强度的平均值。采用与上述方法相同的方法，定位装置可以获取第二栅格(1,2)的反射强度信息和第一栅格(2,8)的反射强度信息的第一偏差
Figure PCTCN2021124206-appb-000023
获取第二栅格(1,3)的反射强度信息和第一栅格(2,9)的反射强度信息的第一偏差
Figure PCTCN2021124206-appb-000024
获取第二栅格(2,1)的反射强度信息和第一栅格(3,7)的反射强度信息的第一偏差
Figure PCTCN2021124206-appb-000025
获取第二栅格(3,3)的反射强度信息和第一栅格(4,9)的反射强度信息的第一偏差
Figure PCTCN2021124206-appb-000026
后续，定位装置可以获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，第二栅格的反射强度信息与第一栅格的反射强度信息的第一总偏差，即第一总偏差
Figure PCTCN2021124206-appb-000027
后续，定位装置可以根据第一总偏差获取局部栅格402的中心位于栅格地图401中黑色的第一栅格中心时，黑色的第一栅格的反射强度信息对应的第一匹配概率。该第一匹配概率r(z|x i,y i,m)可以满足公式：r(z|x i,y i,m)＝μe -A。
可以理解的，定位装置可以采用与上述方法相同的方法，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，第二栅格的反射强度信息与第一栅格的反射强度信息的第一总偏差，根据每个第一总偏差，获取局部栅格402的中心位于栅格地图401中的其他62个第一栅格中的每个第一栅格中心时，每个第一栅格的反射强度信息对应的第一匹配概率。
一种可能的实现方式，若定位装置是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的反射强度信息对应的第一预测概率为预设的初始值，如1与栅格地图中的第一栅格的数量之比，即任意一个第一栅格的反射强度信息对应的第一预测概率满足公式
Figure PCTCN2021124206-appb-000028
其中，
Figure PCTCN2021124206-appb-000029
为任意一个第一栅格的反射强度信息对应的第一预测概率，x i为该第一栅格的中心的横坐标，y i为该第一栅格的中心的纵坐标。
另一种可能的实现方式，若定位装置不是第一次确定待定位装置的位置，则栅格地图中的任意一个第一栅格的反射强度信息对应的第一预测概率满足公式：
Figure PCTCN2021124206-appb-000030
其中，η为常数。r(x 0,y 0|z,m)为最近一次获取的、待定位装置在栅格地图的位置为(x 0,y 0)时的反射强度信息对应的第一后验概率，(x 0,y 0)为定位装置根据待定位装置在上一时刻的位置信息和待定位装置的运动参数，预测的待定位装置在当前时刻在栅 格地图中所处的位置。σ为方差。
S304：定位装置根据第一匹配概率确定待定位装置的位置。
其中，待定位装置的位置为待定位装置在栅格地图中的位置。示例性的，以待定位装置为智能车为例，待定位装置的位置为智能车的质心在栅格地图中的位置。
一种可能的实现方式，定位装置根据不同位置的每种属性信息对应的第一匹配概率确定每种属性信息对应的位置信息，根据每种属性信息对应的位置信息，确定待定位装置的位置。
首先，介绍定位装置根据不同位置的每种属性信息对应的第一匹配概率确定每种属性信息对应的位置信息的过程。
一种可能的实现方式，定位装置根据不同位置的每种属性信息对应的第一匹配概率，以及不同位置的每种属性信息对应的第一预测概率，确定不同位置的每种属性信息对应的第一后验概率，根据不同位置的每种属性信息对应的第一后验概率，确定每种属性信息对应的位置信息。
对于语义信息，栅格地图中的任意一个第一栅格的语义信息对应的第一后验概率为该第一栅格的语义信息对应的第一预测概率与该第一栅格的语义信息对应的第一匹配概率之积。
以上述示例1为例，若局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则栅格地图401中黑色的第一栅格的语义信息对应的第一后验概率满足公式：
Figure PCTCN2021124206-appb-000031
语义信息对应的位置信息满足公式：
Figure PCTCN2021124206-appb-000032
其中，
Figure PCTCN2021124206-appb-000033
为基于语义信息确定的待定位装置在栅格地图中的横坐标，即语义信息对应的待定位装置在栅格地图中的横坐标，x i为栅格地图中的第i个第一栅格的中心的横坐标，p(x i,y i|z,m)为栅格地图中的第i个第一栅格的语义信息对应的第一后验概率，N为栅格地图中包括的第一栅格的数量，
Figure PCTCN2021124206-appb-000034
为基于语义信息确定的待定位装置在栅格地图中的纵坐标，即语义信息对应的待定位装置在栅格地图中的纵坐标，y i为栅格地图中的第i个第一栅格的中心的纵坐标。
对于高度信息，栅格地图中的任意一个第一栅格的高度信息对应的第一后验概率为该第一栅格的高度信息对应的第一预测概率与该第一栅格的高度信息对应的第一匹配概率之积。
以上述示例2为例，若局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则栅格地图401中黑色的第一栅格的高度信息对应的第一后验概率满足公式：
Figure PCTCN2021124206-appb-000035
高度信息对应的位置信息满足公式：
Figure PCTCN2021124206-appb-000036
其中，
Figure PCTCN2021124206-appb-000037
为基于高度信息确定的待定位装置在栅格地图中的横坐标，即高度信息对应的待定位装置在栅格地图中的横坐标，q(x i,y i|z,m)为栅格地图中的第i个第一栅格的高度信息对应的第一后验概率，
Figure PCTCN2021124206-appb-000038
为基于高度信息确定的待定位装置在栅格地图中的纵坐标，即高度信息对应的待定位装置 在栅格地图中的纵坐标。
对于反射强度信息，栅格地图中的任意一个第一栅格的反射强度信息对应的第一后验概率为该第一栅格的反射强度信息对应的第一预测概率与该第一栅格的反射强度信息对应的第一匹配概率之积。
以上述示例3为例，若局部栅格402的中心位于栅格地图401中黑色的第一栅格中心，则栅格地图401中黑色的第一栅格的反射强度信息对应的第一后验概率满足公式：
Figure PCTCN2021124206-appb-000039
反射强度信息对应的位置信息满足公式：
Figure PCTCN2021124206-appb-000040
其中，
Figure PCTCN2021124206-appb-000041
为基于反射强度信息确定的待定位装置在栅格地图中的横坐标，即反射强度信息对应的待定位装置在栅格地图中的横坐标，r(x i,y i|z,m)为栅格地图中的第i个第一栅格的反射强度信息对应的第一后验概率，
Figure PCTCN2021124206-appb-000042
为基于反射强度信息确定的待定位装置在栅格地图中的纵坐标，即反射强度信息对应的待定位装置在栅格地图中的纵坐标。
下面介绍定位装置根据每种属性信息对应的位置信息，确定待定位装置的位置的过程。
一种可能的实现方式，定位装置对每种属性信息对应的位置信息加权求和，得到待定位装置的位置。
可以理解的，在对每种属性信息对应的位置信息加权求和，得到待定位装置的位置的过程中，可以先确定每种属性信息对应的权重。可以理解的，每种属性信息对应的权重可以相同也可以不同，例如，每种属性信息对应的权重都为1。每种属性信息对应的权重可以是预先设置好，存储在定位装置中的，也可以是根据某种算法确定的。
一种可能的实现方式，每种属性信息对应的权重是根据该属性信息对应的第一匹配概率的分布情况确定的。如：栅格地图内该属性信息对应的第一匹配概率的分布越集中，该属性信息对应的权重越高，栅格地图内该属性信息对应的第一匹配概率的分布越不集中，该属性信息对应的权重越低。
作为一种示例，每种属性信息对应的权重是根据该属性信息对应的第一匹配概率的方差确定的。例如，语义信息对应的权重可以满足公式：
Figure PCTCN2021124206-appb-000043
其中，w p为语义信息对应的权重，ξ为常数，u p为语义信息对应的第一匹配概率的方差。高度信息对应的权重可以满足公式：
Figure PCTCN2021124206-appb-000044
其中，w h为高度信息对应的权重，u h为高度信息对应的第一匹配概率的方差。反射强度信息对应的权重可以满足公式：
Figure PCTCN2021124206-appb-000045
其中，w a为反射强度信息对应的权重，u a为反射强度信息对应的第一匹配概率的方差。可以理解的，方差越小，则该方差对应的第一匹配概率的分布越集中，方差越大，则该方差对应的第一匹配概率的分布越不集中。
一种可能的实现方式，待定位装置的位置满足如下公式：
Figure PCTCN2021124206-appb-000046
Figure PCTCN2021124206-appb-000047
其中，X为定位装置确定的待定位装置在栅格地图中的横坐标，Y为定位装置确定的待定位装置在栅格地图中的纵坐标。
可以理解的，为了使得定位装置确定的待定位装置的位置更为准确，定位装置在 确定了每种属性信息对应的位置信息后，可以先对这些位置信息进行筛选，剔除不准确的位置信息，再根据筛选后的位置信息确定待定位装置的位置。
一种可能的实现方式，定位装置在每种属性信息对应的位置信息中确定符合预设规则的至少一种属性信息对应的位置信息，根据至少一种属性信息对应的位置信息，确定待定位装置的位置。
其中，预设规则可以包括以下至少一种规则：将每种属性信息对应的位置信息中，距第一位置的距离小于或等于第一数值的位置信息，确定为至少一种属性信息对应的位置信息；或者，将每种属性信息对应的位置信息中，匹配度大于或等于第二数值的位置信息，确定为至少一种属性信息对应的位置信息。
其中，第一位置为定位装置根据待定位装置在上一时刻的位置信息和待定位装置的运动参数，预测的待定位装置在当前时刻在栅格地图中所处的位置。任意一种属性信息对应的位置信息的匹配度可以是基于ICP算法，将该位置信息在第一环境对应的点云地图中对应的点，与第一环境对应的点云中的点进行匹配的得到的。其中，第一环境对应的点云地图以及第一环境对应的点云的介绍可以参考后续图10所示方法中对应的描述，在此不做赘述。
一种可能的实现方式，若至少一种属性信息的数量大于1，则对至少一种属性信息对应的位置信息加权求和，得到待定位装置的位置；或者，若至少一种属性信息的数量等于1，则将至少一种属性信息对应的位置信息确定为待定位装置的位置。其中，属性信息对应的权重的确定方式如前文所述，在此不做赘述。
需要说明的是，定位装置也可以不确定不同位置的每种属性信息对应的第一后验概率，而是直接根据不同位置的每种属性信息对应的第一匹配概率，确定每种属性信息对应的位置信息，再根据每种属性信息对应的位置信息，确定待定位装置的位置。应理解，属性信息对应的第一后验概率是考虑了属性信息对应的第一预测概率，因此，根据不同位置的每种属性信息对应的第一后验概率确定的待定位装置的位置，比直接根据不同位置的每种属性信息对应的第一匹配概率确定的待定位装置的位置更为准确。基于图3所示的方法，定位装置可以获取第一环境对应的栅格地图，获取第二环境信息对应的局部栅格，根据栅格地图中第一栅格的多种属性信息和局部栅格中第二栅格的多种属性信息对待定位装置进行定位，使得定位装置在定位时不是仅依靠某一种属性来定位，提高了定位的准确性和稳定性。
可以理解的，上述图3所示的方法中，定位装置在每次定位时，都会逐一获取局部栅格与栅格地图中的栅格的第一匹配概率，计算开销较大。在具体应用中，定位装置可以在第一次确定待定位装置的位置时，逐一获取局部栅格与栅格地图中的栅格的匹配概率。而在定位装置不是第一次确定待定位装置的位置的情况下，定位装置可以基于直方图滤波器的方法将局部栅格与栅格地图中的栅格进行匹配。例如，定位装置可以先预测待定位装置在栅格地图的位置，再获取局部栅格与栅格地图中、该位置对应的栅格的匹配概率。如此，可以不用逐一获取局部栅格与栅格地图中的栅格的匹配概率，减少了计算开销。
可选的，如图5所示，在图3所示方法的一种可能的实现方式中，图3所示的方法还可以包括S305。
S305：定位装置根据待定位装置在第一时刻的位姿信息和待定位装置的运动参数，预测待定位装置在第二时刻的位置对应的第一区域。
其中，第一时刻为定位装置最近一次为待定位装置进行定位的时刻。第一时刻的位姿信息用于指示定位装置最近一次确定的待定位装置在栅格地图中的位置和朝向角。待定位装置在栅格地图中的朝向角可以指示待定位装置在栅格地图中的运动方向。第二时刻为当前时刻。第二时刻的位置为当前时刻待定位装置在栅格地图中的位置。也就是说，定位装置可以根据最近一次确定的待定位装置的位姿信息和待定位装置的运动参数，预测当前时刻待定位装置的位置。
其中，运动参数可以是感知模块实时获取的参数。感知模块可以是图1A中的感知装置102，或图1B中的感知模块1112。在这种情况下，运动参数为惯性测量单元(inertial measurement unit，IMU)数据。感知模块也可以是安装在待定位装置上的轮速传感器，在这种情况下，运动参数为轮速计参数。定位装置获取到运动参数后，可以根据运动参数获取待定位装置的角速度、待定位装置的线速度等参数。可选的，定位装置还可以根据运动参数获取待定位装置车轮的转角。
作为一种示例，待定位装置的角速度、待定位装置的线速度和待定位装置车轮的转角可以如图6所示。图6中的线条601和线条602用于表示智能车的两个后轮，线条603和线条604用于表示智能车的两个前轮。待定位装置的角速度ω可以为待定位装置在单位时间内走过的弧度。待定位装置的线速度v可以为智能车两个后轮连线的中心点在单位时间内走过的距离。待定位装置车轮的转角θ可以为智能车的车轮与智能车前轮中心和后轮中心的连线的夹角。
一种可能的实现方式，待定位装置在第一时刻的的位姿(x t1,y t1,θ t1)和预测的待定位装置在第二时刻的位置(x t2,y t2)满足公式：
Figure PCTCN2021124206-appb-000048
其中，x t1为待定位装置在第一时刻在栅格地图中的横坐标。y t1为待定位装置在第一时刻在栅格地图中的纵坐标。x t2为预测的待定位装置在第二时刻在栅格地图中的横坐标。y t2为预测的待定位装置在第二时刻在栅格地图中的纵坐标。v为待定位装置的线速度。ω为待定位装置的角速度。Δt为第二时刻与第一时刻之差。
可以理解的，定位装置在预测待定位装置在第二时刻的位置时可能会出现误差，因此，可以在定位装置预测了待定位装置在第二时刻的位置后，根据预测的待定位装置在第二时刻的位置确定第一区域。后续，定位装置可以获取局部栅格与第一区域中的栅格的第一匹配概率，根据该第一匹配概率确定待定位装置的位置。如此，可以提高定位装置定位的准确性和稳定性，还可以减少计算开销。
作为一种示例，定位装置可以将以预测的待定位装置在第二时刻的位置为中心的一定范围内的区域确定为第一区域。例如，若待定位装置在第一时刻的的位姿和预测的待定位装置在第二时刻的位置如图7所示，则第一区域可以为图7中的阴影部分。
可以理解的，定位装置确定了第一区域后，可以不需要按照上述图3所示的方法获取第一匹配概率，根据该第一匹配概率确定待定位装置的位置。定位装置可以获取局部栅格位于第一区域中的栅格时，局部栅格与上个地图中的栅格的第一匹配概率， 即第一区域中的栅格的第一匹配概率，根据第一区域中的栅格的第一匹配概率确定待定位装置的位置。下面进行具体阐述。
一种可能的实现方式，定位装置根据局部栅格中的栅格的属性信息和第一区域中的栅格的属性信息，获取不同位置的每种属性信息对应的第一匹配概率。具体的，定位装置获取局部栅格位于第一区域中的第一栅格的情况下，局部栅格中的第二栅格的第一属性信息与第二栅格对应的第一栅格的第一属性信息之间的第一偏差，根据第一偏差，获取第一区域中的第一栅格的第一属性信息对应的第一匹配概率。其中，第二栅格的第一属性信息为第二栅格的属性信息中的任意一个属性信息，第一栅格的第一属性信息与第二栅格的第一属性信息的属性相同。上述具体过程与S303中定位装置获取栅格地图中的第一栅格的第一属性信息对应的第一匹配概率的过程类似，因此可以参考S303中对于的描述，在此不做赘述。
一种可能的实现方式，定位装置还可以获取第一区域中的第一栅格的每种属性信息对应的第一预测概率，以便定位装置根据第一区域中的第一栅格的每种属性信息对应的第一匹配概率和第一预测概率，确定第一区域中的第一栅格的每种属性信息对应的第一后验概率，进而根据第一区域中的第一栅格的每种属性信息对应的第一后验概率确定待定位装置的位置。定位装置获取第一区域中的第一栅格的每种属性信息对应的第一预测概率的过程与上述S303中定位装置获取栅格地图中的第一栅格的每种属性信息对应的第一预测概率的过程类似，因此可以参考上述S303中对应的描述，在此不做赘述。
一种可能的实现方式，定位装置根据第一区域中的第一栅格的每种属性信息对应的第一匹配概率确定每种属性信息对应的位置信息，根据每种属性信息对应的位置信息，确定待定位装置的位置。具体的，定位装置根据第一区域中的第一栅格的每种属性信息对应的第一匹配概率和第一预测概率，确定第一区域中的第一栅格的每种属性信息对应的第一后验概率，根据第一区域中的第一栅格的每种属性信息对应的第一后验概率，确定每种属性信息对应的位置信息，根据每种属性信息对应的位置信息，确定待定位装置的位置。上述过程与上述S304中定位装置确定待定位装置的位置的过程类似，因此可以参考上述S304中对应的描述，在此不做赘述。
可以理解的，为了使得定位装置确定的待定位装置的位置更为准确，定位装置在确定了每种属性信息对应的位置信息后，可以先对这些位置信息进行筛选，剔除不准确的位置信息，再根据筛选后的位置信息确定待定位装置的位置。具体的，可以参考上述S304中对应的描述，在此不做赘述。
根据上述图3或图5所示的方法，定位装置可以对待定位装置定位，获取待定位装置的位置。在具体应用中，定位装置还可以获取待定位装置的朝向角。
可选的，如图8所示，在图3所示方法的一种可能的实现方式中，图3所示的方法还包括S306-S307。
S306：定位装置获取第二匹配概率。
其中，第二匹配概率为在局部栅格位于待定位装置的位置，并且局部栅格旋转不同角度的情况下，局部栅格与栅格地图中的栅格的匹配概率。局部栅格的不同旋转角对应的待定位装置的不同朝向角。
例如，如图9所示，可以将局部栅格的中心位于待定位装置的位置，如图9中的局部栅格901，并以待定位装置的位置为中心旋转局部栅格，可以得到不同旋转角的局部栅格，如图9中的局部栅格902-局部栅格904。局部栅格901的旋转角对应的待定位装置的朝向角为箭头905对应的朝向角，局部栅格902的旋转角对应的待定位装置的朝向角为箭头906对应的朝向角，局部栅格903的旋转角对应的待定位装置的朝向角为箭头907对应的朝向角，局部栅格904的旋转角对应的待定位装置的朝向角为箭头908对应的朝向角。后续，定位装置可以获取不同旋转角的局部栅格与栅格地图中的栅格的第二匹配概率，例如，局部栅格901与栅格地图中的栅格的第二匹配概率，局部栅格902与栅格地图中的栅格的第二匹配概率，局部栅格903与栅格地图中的栅格的第二匹配概率，局部栅格904与栅格地图中的栅格的第二匹配概率。
其中，局部栅格的旋转角为旋转后的局部栅格局与初始位于待定位装置的位置的局部栅格之间的夹角。以图9所示为例，局部栅格902的旋转角为局部栅格902和局部栅格901之间的夹角，局部栅格903的旋转角为局部栅格903和局部栅格901之间的夹角。本申请实施例中，局部栅格相邻两次旋转的旋转角之差可以相同也可以不同，不予限制。
一种可能的实现方式，定位装置获取局部栅格位于待定位装置的位置的情况下，局部栅格对应的多个旋转角的第二匹配概率。
在一种可能的设计中，将局部栅格的中心位于待定位装置的位置，并将局部栅格以待定位装置的位置为中心，旋转一圈，得到多个旋转角。例如，多个旋转角可以为0°，45°，90°，135°，180°，225°，270°和315°。
在另一种可能的设计中，若定位装置不是第一次确定待定位装置的朝向角，则定位装置可以先根据待定位装置在第三时刻的朝向角和待定位装置的运动参数，预测待定位装置在第二时刻的朝向角，根据该第二时刻的朝向角确定多个旋转角。在这种情况下，不需要将局部栅格以待定位装置的位置为中心，旋转一圈来得到多个旋转角，而是将局部栅格以待定位装置的位置为中心，在第二时刻的朝向角附近旋转来得到多个旋转角。例如，多个旋转角中，每个旋转角与第二时刻的朝向角之间的夹角小于或等于第一角度。如此，既可以降低定位装置的计算开销，又可以提高确定的待定位装置的朝向角的精度。
其中，第三时刻为定位装置最近一次为待定位装置确定朝向角的时刻。第三时刻的朝向角为定位装置最近一次确定的待定位装置在栅格地图中的朝向角。第三时刻和第一时刻可以相同也可以不同。待定位装置在栅格地图中的朝向角可以指示待定位装置在栅格地图中的运动方向。第二时刻为当前时刻。第二时刻的朝向角为当前时刻待定位装置在栅格地图中的朝向角。也就是说，定位装置可以根据最近一次确定的待定位装置的朝向角和待定位装置的运动参数，预测当前时刻待定位装置在栅格地图的朝向角。其中，运动参数的介绍可以参考上述S305中对应的描述，在此不做赘述。
可以理解的，定位装置获取到运动参数后，可以根据运动参数获取待定位装置的角速度等参数。可选的，定位装置还可以根据运动参数获取待定位装置的线速度、待定位装置车轮的转角等。
一种可能的实现方式，待定位装置在第三时刻的的朝向角θ t3和预测的待定位装置 在第二时刻的朝向角θ t2满足公式：θ t2＝θ t3+ωΔt。其中，ω为待定位装置的角速度。Δt为第二时刻与第三时刻之差。
可以理解的，上述第一栅格或第二栅格对应有多种属性信息。定位装置可以基于每种属性信息确定每种属性信息对应的第二匹配概率。示例性的，定位装置根据局部栅格中的栅格的属性信息和栅格地图中的栅格的属性信息，获取多个旋转角对应的局部栅格的每种属性信息对应的第二匹配概率。
一种可能的实现方式，定位装置根据第一旋转角对应的局部栅格中的栅格的属性信息和栅格地图中的栅格的属性信息，获取第一旋转角的每种属性信息对应的第二匹配概率。其中，第一旋转角为多个旋转角中的任意一个旋转角，第一旋转角的每种属性信息对应的第二匹配概率为第一旋转角对应的局部栅格中的栅格的每种属性信息，与栅格地图中的栅格的每种属性信息的匹配概率。具体的，定位装置获取局部栅格位于待定位装置的位置，并且局部栅格旋转第一旋转角的情况下，局部栅格中的每个第二栅格的第一属性信息，与第二栅格对应的第一栅格的第一属性信息之间的第二偏差，根据第二偏差，获取第一旋转角的第一属性信息对应的第二匹配概率。其中，第二栅格的第一属性信息为第二栅格的属性信息中的任意一个属性信息，第一栅格的第一属性信息与第二栅格的第一属性信息的属性相同。可以理解的，定位装置获取第二偏差，根据第二偏差获取第一旋转角的第一属性信息对应的第二匹配概率的过程，与上述S303中定位装置获取第一偏差，根据第一偏差获取栅格地图中的第一栅格的第一属性信息对应的第一匹配概率的过程类似，可以参考上述S303中对应的描述，在此不做赘述。
一种可能的实现方式，定位装置还可以获取第一旋转角的每种属性信息对应的第二预测概率，以便定位装置根据第一旋转角的每种属性信息对应的第二匹配概率和第二预测概率，确定第一旋转角的每种属性信息对应的第二后验概率，进而根据第一旋转角的每种属性信息对应的第二后验概率确定待定位装置的朝向角。
在一种可能的设计中，若定位装置是第一次确定待定位装置的位置，则第一旋转角的每种属性信息对应的第二预测概率为预设的初始值，如1/360。
在另一种可能的设计中，若定位装置不是第一次确定待定位装置的位置，则第一旋转角的第一属性信息对应的第二预测概率满足公式：
Figure PCTCN2021124206-appb-000049
其中，
Figure PCTCN2021124206-appb-000050
为第一旋转角的属性信息对应的第二预测概率。η为常数。k(θ 0|z,m)为最近一次获取的、待定位装置在栅格地图的朝向角为θ 0时的第一属性信息对应的第二后验概率，θ 0为定位装置根据待定位装置在上一时刻的朝向角和待定位装置的运动参数，预测的待定位装置在当前时刻在栅格地图中的朝向角。θ i为第一旋转角对应的待定位装置的朝向角。σ为方差。
S307：定位装置根据第二匹配概率确定待定位装置的朝向角。
一种可能的实现方式，定位装置根据每个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息，根据每种属性信息对应的朝向角信息，确定待定位装置的朝向角。具体的，定位装置根据每个旋转角的每种属性信息对应的第二匹配概率和第二预测概率，确定每个旋转角的每种属性信息对应的第二后验概率， 根据每个旋转角的每种属性信息对应的第二后验概率，确定每种属性信息对应的朝向角信息，根据每种属性信息对应的朝向角信息，确定待定位装置的朝向角。
作为一种示例，第一旋转角的第一属性信息对应的第二后验概率满足公式：
Figure PCTCN2021124206-appb-000051
其中，k(θ i|z,m)为第一旋转角的第一属性信息对应的第二后验概率，
Figure PCTCN2021124206-appb-000052
为第一旋转角的第一属性信息对应的第二预测概率，k(z|θ i,m)为第一旋转角的第一属性信息对应的第二匹配概率。第一属性信息对应的朝向角信息满足公式：
Figure PCTCN2021124206-appb-000053
其中，
Figure PCTCN2021124206-appb-000054
为第一属性信息对应的朝向角信息指示的朝向角，即基于第一属性信息确定的待定位装置的朝向角，θ i为多个旋转角中的第i个旋转角对应的待定位装置的朝向角，j为多个旋转角的数量。
一种可能的实现方式，定位装置对所述每种属性信息对应的朝向角信息加权求和，得到待定位装置的朝向角。其中，每种属性信息对应的权重的介绍可以参考上述S304中对应的描述，在此不做赘述。
作为一种示例，待定位装置的朝向角满足如下公式：
Figure PCTCN2021124206-appb-000055
其中，θ为定位装置确定的待定位装置的朝向角，w p为语义信息对应的权重，w h为高度信息对应的权重，w a为反射强度信息对应的权重。
Figure PCTCN2021124206-appb-000056
为基于语义信息确定的待定位装置的朝向角，
Figure PCTCN2021124206-appb-000057
为基于高度信息确定的待定位装置的朝向角，
Figure PCTCN2021124206-appb-000058
为基于反射强度信息确定的待定位装置的朝向角。
需要说明的是，定位装置也可以不确定多个旋转角的每种属性信息对应的第二后验概率，而是直接根据多个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息，再根据每种属性信息对应的朝向角信息，确定待定位装置的朝向角。应理解，属性信息对应的第二后验概率是考虑了属性信息对应的第二预测概率，因此，根据多个旋转角的每种属性信息对应的第二后验概率确定的待定位装置的朝向角，比直接根据多个旋转角的每种属性信息对应的第二匹配概率确定的待定位装置的朝向角更为准确。
可以理解的，上述S301-S307中的定位装置的动作可以由图2所示的定位装置20中的处理器201调用存储器203中存储的应用程序代码来执行，本申请实施例对此不做任何限制。
可选的，如图10所示，在图3所示方法的一种可能的实现方式中，S301可以包括S3011和S3012。
S3011：定位装置获取第一环境对应的点云的信息。
其中，第一环境对应的点云的信息可以用于指示第一环境对应的点云中点的位置信息，如第一环境对应的点云中的点在三维地图中的坐标。该三维地图可以是点云地图。
可选的，第一环境对应的点云的信息还用于指示第一环境对应的点云中点的反射强度。
作为一种示例，定位装置通过感知装置获取第一环境对应的点云的信息。感知装 置可以部署在定位装置上，也可以不部署在定位装置上。以感知装置为激光雷达为例，激光雷达可以在第一环境中移动，并发射激光，采集第一环境中的点云信息。将第一环境的点云信息发送给定位装置，定位装置基于SLAM算法，如轻量级和地面优化的激光雷达里程计和建图(lightweight and ground optimized lidar odometry and mapping，Lego-Loam)算法，获取第一环境对应的点云地图。其中，第一环境中的点云为第一环境中的物体反射激光雷达得到的点云。第一环境对应的点云中点的反射强度为该点对激光的反射强度。
一种可能的实现方式，在通过激光雷达采集点云信息时，可以间隔一定的距离采集一次点云信息。
可以理解的，通过激光雷达采集的点云，往往会较为密集，过多的点云数量会对后续分割工作带来困难。因此，可以通过点云库中的体素栅格滤波器对每次采集的点云进行降采样。体素栅格滤波器可以达到降采样的同时不破坏点云本身几何结构的功能，经过降采样，可以减少每次采集的点云中的点的数量，并同时保持点云的形状特征。
S3012：定位装置根据第一环境对应的点云的信息，获取第一环境对应的栅格地图。
一种可能的实现方式，定位装置将第一环境对应的点云地图进行投影(如向地面投影)以获取第一环境对应的二维地图，将二维地图划分为多个第一栅格，以得到第一环境对应的栅格地图。其中，栅格地图中的每个点的高度即为该点在点云地图中的竖坐标(z坐标)的值。栅格地图中的每个点的反射强度即为该点对激光的反射强度。
一种可能的实现方式，第一栅格的语义信息可以根据第一栅格中的点的语义信息确定。第一栅格中的点的语义信息为该点所属的物体(或该点所对应的物体)为多种物体中的每种物体的概率。
作为一种示例，第一栅格的语义信息满足公式：p m[i]＝λcount(label i) 1/M。其中，p m[i]为第一栅格表示多种物体中的任意一种物体的概率，λ为常数，count(label i) 1为该第一栅格内属于该任意一种物体的点的数量，M为第一栅格中点的数量。
一种可能的实现方式，将第一栅格中的点输入到语义模型中，可以得到该点所属的物体(或该点所对应的物体)为多种物体中的每种物体的概率。
一种可能的实现方式，语义模型可以是存储在定位装置中的。语义模型也可以是定位装置在S3012之前训练得到的。例如，定位装置可以根据提前标注好的多种物体，使用PointNet进行训练，得到语义模型。
其中，上述S3011-S3012中的定位装置的动作可以由图2所示的定位装置20中的处理器201调用存储器203中存储的应用程序代码来执行，本申请实施例对此不做任何限制。
可选的，如图11所示，在图3所示方法的一种可能的实现方式中，S302可以包括S3021和S3022。
S3021：定位装置获取第二环境对应的点云的信息。
其中，第二环境对应的点云的信息可以用于指示第二环境对应的点云中点的位置信息，如第二环境对应的点云中的点在三维地图中的坐标。该三维地图可以是点云地图。
可选的，第二环境对应的点云的信息还用于指示第二环境对应的点云中点的反射强度。
作为一种示例，待定位装置上部署有感知装置，通过感知装置获取第二环境对应的点云的信息。以感知装置为激光雷达为例，待定位装置可以在第一环境中移动，当移动到第二环境时，通过激光雷达发射激光，采集第二环境中的点云信息。若待定位装置与定位装置相同，则待定位装置基于SLAM算法，如Lego-Loam算法，获取第二环境对应的点云地图。若待定位装置与定位装置不同，则待定位装置向定位装置发送第二环境中的点云信息，定位装置基于SLAM算法，如Lego-Loam算法，获取第二环境对应的点云地图。其中，第二环境中的点云为第二环境中的物体反射激光雷达得到的点云。第二环境对应的点云中点的反射强度为该点对激光的反射强度。
一种可能的实现方式，在通过激光雷达采集点云信息时，可以间隔一定的距离采集一次点云信息。
可以理解的，通过激光雷达采集的点云，往往会较为密集，过多的点云数量会对后续分割工作带来困难。因此，可以通过点云库中的体素栅格滤波器对每次采集的点云进行降采样。
S3022：定位装置根据第二环境对应的点云的信息，获取第二环境对应的局部栅格。
一种可能的实现方式，定位装置将第二环境对应的点云地图进行投影(如向地面投影)以获取第二环境对应的二维地图，将二维地图划分为多个第二栅格，以得到第二环境对应的局部栅格。其中，局部栅格中的每个点的高度即为该点在点云地图中的竖坐标(z坐标)的值。局部栅格中的每个点的反射强度即为该点对激光的反射强度。
一种可能的实现方式，第二栅格的语义信息可以根据第二栅格中的点的语义信息确定。第二栅格中的点的语义信息为该点所属的物体(或该点所对应的物体)为多种物体中的每种物体的概率。
作为一种示例，第二栅格的语义信息满足公式：p[i]＝λcount(label i) 2/W。其中，p[i]为第二栅格表示多种物体中的任意一种物体的概率，λ为常数，count(label i) 2为该第二栅格内属于该任意一种物体的点的数量，W为第二栅格中点的数量。
一种可能的实现方式，将第二栅格中的点输入到语义模型中，可以得到该点所属的物体(或该点所对应的物体)为多种物体中的每种物体的概率。语义模型的介绍可以参考上述S3012中对应的描述，在此不做赘述。
其中，上述S3021-S3022中的定位装置的动作可以由图2所示的定位装置20中的处理器201调用存储器203中存储的应用程序代码来执行，本申请实施例对此不做任何限制。
可以理解的，以上各个实施例中，由定位装置实现的方法和/或步骤，也可以由可用于定位装置的部件(例如芯片或者电路)实现，不予限制。
为了验证本申请实施例提供的几种定位方法的定位精度，采用方法1、方法2和方法4分别对4个位置的待定位装置进行定位，并统计了不同时刻的50次定位结果，统计后的定位结果如表1所示。表1中，对于位置1，即图12所示的待定位装置的实际位置，在50次定位中，方法1对待定位装置进行定位时的定位误差小于10cm的比例为58.9％，方法2对待定位装置进行定位时的定位误差小于10cm的比例为68.1％， 方法4对待定位装置进行定位时的定位误差小于10cm的比例为92.0％。图12示出了，采用方法1进行定位的定位结果和采用方法4进行定位的定位结果。由于停车场的场景较为相似，所以方法1出现的误判，导致定位不准确，而方法4的定位结果与待定位装置的实际位置相符。对于位置2，在50次定位中，方法1对待定位装置进行定位时的定位误差小于10cm的比例为76.2％，方法2对待定位装置进行定位时的定位误差小于10cm的比例为86.3％，方法4对待定位装置进行定位时的定位误差小于10cm的比例为93.9％。对于位置3，在50次定位中，方法1对待定位装置进行定位时的定位误差小于10cm的比例为56.3％，方法2对待定位装置进行定位时的定位误差小于10cm的比例为66.8％，方法4对待定位装置进行定位时的定位误差小于10cm的比例为91.3％。对于位置4，在50次定位中，方法1对待定位装置进行定位时的定位误差小于10cm的比例为67.1％，方法2对待定位装置进行定位时的定位误差小于10cm的比例为85％，方法4对待定位装置进行定位时的定位误差小于10cm的比例为94.2％。可以看出，方法4的定位精度明显优于方法1和方法2。
表1
  位置1 位置2 位置3 位置4
方法1 58.9％ 76.2％ 56.3％ 67.1％
方法2 68.1％ 86.3％ 66.8％ 85％
方法4 92.0％ 93.9％ 91.3％ 94.2％
可以理解的，本申请实施例中的属性信息除了可以用于定位之外，还可以用于自动驾驶的环境感知。例如，可以将环境中不易改变的特征提取出来，作为属性信息，对环境进行建模，或者可以根据本申请实施例中获取语义信息的过程识别环境中的物体，实现对环境中动态物体的运动估计，从而为自动驾驶的路径规划与避障提供条件。
本申请实施例还提供了定位装置，该定位装置可以为上述方法实施例中的定位装置，或者包含上述定位装置的装置，或者为可用于上述定位装置的部件。可以理解的是，上述定位装置为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例的单元及算法操作，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对定位装置进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
比如，以采用集成的方式划分各个功能模块的情况下，图13示出了一种定位装置130的结构示意图。定位装置130包括获取模块1301和处理模块1302。
示例性地，定位装置130用于实现上述定位装置的功能。定位装置130例如为图3所示的实施例，图5所示的实施例，图8所示的实施例，图10所示的实施例或图11 所示的实施例所述的终端。
其中，获取模块1301，用于获取第一环境对应的栅格地图，栅格地图包括多个第一栅格，第一栅格包括多个点、多个点的位置信息和第一栅格的属性信息，第一栅格的属性信息包括第一栅格的语义信息和第一栅格的高度信息，第一栅格的语义信息用于指示第一栅格表示多种物体中的每种物体的概率，第一栅格的高度信息用于指示第一栅格包括的点的高度。
获取模块1301，还用于获取第二环境对应的局部栅格，第二环境包括在第一环境中，局部栅格包括至少一个第二栅格，第二栅格包括多个点和第二栅格的属性信息，第二栅格的属性信息包括第二栅格的语义信息和第二栅格的高度信息，第二栅格的语义信息用于指示第二栅格表示多种物体中的每种物体的概率，第二栅格的高度信息用于指示第二栅格包括的点的高度。
获取模块1301，还用于获取第一匹配概率，第一匹配概率为局部栅格位于栅格地图中的不同位置时，局部栅格与栅格地图中的栅格的匹配概率。
处理模块1302，用于根据第一匹配概率确定待定位装置的位置，待定位装置位于第二环境中。
在一种可能的实现方式中，第一栅格的属性信息还包括第一栅格的反射强度信息，第一栅格的反射强度信息用于指示第一栅格包括的点的反射强度；第二栅格的属性信息还包括第二栅格的反射强度信息，第二栅格的反射强度信息用于指示第二栅格包括的点的反射强度。
在一种可能的实现方式中，处理模块1302，还用于根据待定位装置在第一时刻的位姿信息和待定位装置的运动参数，预测待定位装置在第二时刻的位置对应的第一区域。
在一种可能的实现方式中，获取模块1301，具体用于获取局部栅格位于第一区域中的栅格时，局部栅格与栅格地图中的栅格的第一匹配概率。
在一种可能的实现方式中，获取模块1301，具体用于根据局部栅格中的栅格的属性信息和第一区域中的栅格的属性信息，获取第一区域中第一栅格的每种属性信息对应的第一匹配概率。
在一种可能的实现方式中，获取模块1301，具体用于获取局部栅格位于第一区域中的第一栅格的情况下，局部栅格中的第二栅格的第一属性信息与第二栅格对应的第一栅格的第一属性信息之间的第一偏差，第二栅格的第一属性信息为第二栅格的属性信息中的任意一个属性信息，第一栅格的第一属性信息与第二栅格的第一属性信息的属性相同；获取模块1301，还具体用于根据第一偏差，获取第一区域中的第一栅格的第一属性信息对应的第一匹配概率。
在一种可能的实现方式中，处理模块1302，具体用于根据第一区域中第一栅格的每种属性信息对应的第一匹配概率，确定每种属性信息对应的位置信息；处理模块1302，还具体用于根据每种属性信息对应的位置信息，确定待定位装置的位置。
在一种可能的实现方式中，处理模块1302，还具体用于在每种属性信息对应的位置信息中确定符合预设规则的至少一种属性信息对应的位置信息；处理模块1302，还具体用于若至少一种属性信息的数量大于1，则对至少一种属性信息对应的位置信息 加权求和，得到待定位装置的位置；或者，若至少一种属性信息的数量等于1，则将至少一种属性信息对应的位置信息确定为待定位装置的位置。
在一种可能的实现方式中，获取模块1301，还用于获取第二匹配概率，第二匹配概率为在局部栅格位于待定位装置的位置，并且局部栅格旋转不同角度的情况下，局部栅格与栅格地图中的栅格的匹配概率；处理模块1302，还用于根据第二匹配概率确定待定位装置的朝向角。
在一种可能的实现方式中，处理模块1302，还用于根据待定位装置在第三时刻的朝向角和待定位装置的运动参数，预测待定位装置在第二时刻的朝向角。
在一种可能的实现方式中，获取模块1301，具体用于获取局部栅格位于待定位装置的位置的情况下，局部栅格对应的多个旋转角的第二匹配概率，多个旋转角中，每个旋转角与预测待定位装置在第二时刻的朝向角之间的夹角小于或等于第一角度。
在一种可能的实现方式中，获取模块1301，具体用于根据第一旋转角对应的局部栅格中的栅格的属性信息和栅格地图中的栅格的属性信息，获取第一旋转角的每种属性信息对应的第二匹配概率，第一旋转角为多个旋转角中的任意一个旋转角。
在一种可能的实现方式中，获取模块1301，具体用于获取局部栅格位于待定位装置的位置，并且局部栅格旋转第一旋转角的情况下，局部栅格中的每个第二栅格的第一属性信息，与第二栅格对应的第一栅格的第一属性信息之间的第二偏差，第二栅格的第一属性信息为第二栅格的属性信息中的任意一个属性信息，第一栅格的第一属性信息与第二栅格的第一属性信息的属性相同；获取模块1301，还具体用于根据第二偏差，获取第一旋转角的第一属性信息对应的第二匹配概率。
在一种可能的实现方式中，处理模块1302，具体用于根据每个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息；处理模块1302，还具体用于对每种属性信息对应的朝向角信息加权求和，得到待定位装置的朝向角。
在一种可能的实现方式中，获取模块1301，具体用于获取第一环境对应的点云的信息，第一环境对应的点云的信息用于指示第一环境对应的点云中点的位置信息；获取模块1301，具体用于根据第一环境对应的点云的信息，获取第一环境对应的栅格地图。
在一种可能的实现方式中，第一环境对应的点云的信息还用于指示第一环境对应的点云中点的反射强度。
在一种可能的实现方式中，获取模块1301，具体用于获取第二环境对应的点云的信息，第二环境对应的点云的信息用于指示第二环境对应的点云中点的位置信息；获取模块1301，还具体用于根据第二环境对应的点云的信息，获取第二环境对应的局部栅格。
在一种可能的实现方式中，第二环境对应的点云的信息还用于指示第二环境对应的点云中点的反射强度。
当用于实现上述定位装置的功能时，关于定位装置130所能实现的其他功能，可参考图3所示的实施例，图5所示的实施例，图8所示的实施例，图10所示的实施例或图11所示的实施例的相关介绍，不多赘述。
在一个简单的实施例中，本领域的技术人员可以想到定位装置130可以采用图2 所示的形式。比如，图2中的处理器201可以通过调用存储器203中存储的计算机执行指令，使得定位装置130执行上述方法实施例中所述的方法。
示例性的，图13中的获取模块1301和处理模块1302的功能/实现过程可以通过图2中的处理器201调用存储器203中存储的计算机执行指令来实现。
需要说明的是，以上模块或单元的一个或多个可以软件、硬件或二者结合来实现。当以上任一模块或单元以软件实现的时候，所述软件以计算机程序指令的方式存在，并被存储在存储器中，处理器可以用于执行所述程序指令并实现以上方法流程。该处理器可以内置于SoC(片上系统)或ASIC，也可是一个独立的半导体芯片。该处理器内处理用于执行软件指令以进行运算或处理的核外，还可进一步包括必要的硬件加速器，如现场可编程门阵列(field programmable gate array，FPGA)、PLD(可编程逻辑器件)、或者实现专用逻辑运算的逻辑电路。
当以上模块或单元以硬件实现的时候，该硬件可以是CPU、微处理器、数字信号处理(digital signal processing，DSP)芯片、微控制单元(microcontroller unit，MCU)、人工智能处理器、ASIC、SoC、FPGA、PLD、专用数字电路、硬件加速器或非集成的分立器件中的任一个或任一组合，其可以运行必要的软件或不依赖于软件以执行以上方法流程。
可选的，本申请实施例还提供了一种芯片系统，包括：至少一个处理器和接口，该至少一个处理器通过接口与存储器耦合，当该至少一个处理器执行存储器中的计算机程序或指令时，使得上述任一方法实施例中的方法被执行。在一种可能的实现方式中，该芯片系统还包括存储器。可选的，该芯片系统可以由芯片构成，也可以包含芯片和其他分立器件，本申请实施例对此不作具体限定。
可选的，本申请实施例还提供了一种计算机可读存储介质。上述方法实施例中的全部或者部分流程可以由计算机程序来指令相关的硬件完成，该程序可存储于上述计算机可读存储介质中，该程序在执行时，可包括如上述各方法实施例的流程。计算机可读存储介质可以是前述任一实施例的定位装置的内部存储单元，例如定位装置的硬盘或内存。上述计算机可读存储介质也可以是上述定位装置的外部存储设备，例如上述定位装置上配备的插接式硬盘，智能存储卡(smart media card，SMC)，安全数字(secure digital，SD)卡，闪存卡(flash card)等。进一步地，上述计算机可读存储介质还可以既包括上述定位装置的内部存储单元也包括外部存储设备。上述计算机可读存储介质用于存储上述计算机程序以及上述定位装置所需的其他程序和数据。上述计算机可读存储介质还可以用于暂时地存储已经输出或者将要输出的数据。
可选的，本申请实施例还提供了一种计算机程序产品。上述方法实施例中的全部或者部分流程可以由计算机程序来指令相关的硬件完成，该程序可存储于上述计算机程序产品中，该程序在执行时，可包括如上述各方法实施例的流程。
可选的本申请实施例还提供了一种计算机指令。上述方法实施例中的全部或者部分流程可以由计算机指令来指令相关的硬件(如计算机、处理器、接入网设备、移动性管理网元或会话管理网元等)完成。该程序可被存储于上述计算机可读存储介质中或上述计算机程序产品中。
通过以上的实施方式的描述，所属领域的技术人员可以清楚地了解到，为描述的 方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个装置，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是一个物理单元或多个物理单元，即可以位于一个地方，或者也可以分布到多个不同地方。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何在本申请揭露的技术范围内的变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (38)

  1. 一种定位方法，其特征在于，所述方法包括：
    获取第一环境对应的栅格地图，所述栅格地图包括多个第一栅格，所述第一栅格包括多个点、所述多个点的位置信息和所述第一栅格的属性信息，所述第一栅格的属性信息包括所述第一栅格的语义信息和所述第一栅格的高度信息，所述第一栅格的语义信息用于指示所述第一栅格表示多种物体中的每种物体的概率，所述第一栅格的高度信息用于指示所述第一栅格包括的点的高度；
    获取第二环境对应的局部栅格，所述第二环境包括在所述第一环境中，所述局部栅格包括至少一个第二栅格，所述第二栅格包括多个点和所述第二栅格的属性信息，所述第二栅格的属性信息包括所述第二栅格的语义信息和所述第二栅格的高度信息，所述第二栅格的语义信息用于指示所述第二栅格表示多种物体中的每种物体的概率，所述第二栅格的高度信息用于指示所述第二栅格包括的点的高度；
    获取第一匹配概率，所述第一匹配概率为所述局部栅格位于所述栅格地图中的不同位置时，所述局部栅格与所述栅格地图中的栅格的匹配概率；
    根据所述第一匹配概率确定待定位装置的位置，所述待定位装置位于所述第二环境中。
  2. 根据权利要求1所述的方法，其特征在于，
    所述第一栅格的属性信息还包括所述第一栅格的反射强度信息，所述第一栅格的反射强度信息用于指示所述第一栅格包括的点的反射强度；
    所述第二栅格的属性信息还包括所述第二栅格的反射强度信息，所述第二栅格的反射强度信息用于指示所述第二栅格包括的点的反射强度。
  3. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    根据所述待定位装置在第一时刻的位姿信息和所述待定位装置的运动参数，预测所述待定位装置在第二时刻的位置对应的第一区域。
  4. 根据权利要求3所述的方法，其特征在于，所述获取第一匹配概率，包括：
    获取所述局部栅格位于所述第一区域中的栅格时，所述局部栅格与所述栅格地图中的栅格的第一匹配概率。
  5. 根据权利要求4所述的方法，其特征在于，所述获取所述局部栅格位于所述第一区域中的栅格时，所述局部栅格与所述栅格地图中的栅格的第一匹配概率，包括：
    根据所述局部栅格中的栅格的属性信息和所述第一区域中的栅格的属性信息，获取第一区域中第一栅格的每种属性信息对应的第一匹配概率。
  6. 根据权利要求5所述的方法，其特征在于，根据所述局部栅格中的栅格的属性信息和所述第一区域中的栅格的属性信息，获取第一区域中第一栅格的第一属性信息对应的第一匹配概率，包括：
    获取所述局部栅格位于所述第一区域中的第一栅格的情况下，所述局部栅格中的第二栅格的第一属性信息与所述第二栅格对应的第一栅格的第一属性信息之间的第一偏差，所述第二栅格的第一属性信息为所述第二栅格的属性信息中的任意一个属性信息，所述第一栅格的第一属性信息与所述第二栅格的第一属性信息的属性相同；
    根据所述第一偏差，获取所述第一区域中的第一栅格的第一属性信息对应的第一 匹配概率。
  7. 根据权利要求5或6所述的方法，其特征在于，所述根据所述第一匹配概率确定待定位装置的位置，包括：
    根据所述第一区域中第一栅格的每种属性信息对应的第一匹配概率，确定每种属性信息对应的位置信息；
    根据所述每种属性信息对应的位置信息，确定所述待定位装置的位置。
  8. 根据权利要求7所述的方法，其特征在于，所述根据所述第一匹配概率确定所述待定位装置的位置，还包括：
    在所述每种属性信息对应的位置信息中确定符合预设规则的至少一种属性信息对应的位置信息；
    所述根据所述每种属性信息对应的位置信息，确定所述待定位装置的位置，包括：
    若所述至少一种属性信息的数量大于1，则对所述至少一种属性信息对应的位置信息加权求和，得到所述待定位装置的位置；或者，
    若所述至少一种属性信息的数量等于1，则将所述至少一种属性信息对应的位置信息确定为所述待定位装置的位置。
  9. 根据权利要求1-8中任一项所述的方法，其特征在于，所述方法还包括：
    获取第二匹配概率，所述第二匹配概率为在所述局部栅格位于所述待定位装置的位置，并且所述局部栅格旋转不同角度的情况下，所述局部栅格与所述栅格地图中的栅格的匹配概率；
    根据所述第二匹配概率确定所述待定位装置的朝向角。
  10. 根据权利要求9所述的方法，其特征在于，所述方法还包括：
    根据所述待定位装置在第三时刻的朝向角和所述待定位装置的运动参数，预测所述待定位装置在第二时刻的朝向角。
  11. 根据权利要求10所述的方法，其特征在于，所述获取所述第二匹配概率，包括：
    获取所述局部栅格位于所述待定位装置的位置的情况下，所述局部栅格对应的多个旋转角的第二匹配概率，所述多个旋转角中，每个旋转角与所述预测所述待定位装置在第二时刻的朝向角之间的夹角小于或等于第一角度。
  12. 根据权利要求11所述的方法，其特征在于，获取所述局部栅格位于所述待定位装置的位置的情况下，第一旋转角的第二匹配概率，包括：
    根据所述第一旋转角对应的局部栅格中的栅格的属性信息和所述栅格地图中的栅格的属性信息，获取所述第一旋转角的每种属性信息对应的第二匹配概率，所述第一旋转角为所述多个旋转角中的任意一个旋转角。
  13. 根据权利要求12所述的方法，其特征在于，根据所述第一旋转角对应的局部栅格中的栅格的属性信息和所述栅格地图中的栅格的属性信息，获取所述第一旋转角的第一属性信息对应的第二匹配概率，包括：
    获取所述局部栅格位于所述待定位装置的位置，并且所述局部栅格旋转第一旋转角的情况下，所述局部栅格中的每个第二栅格的第一属性信息，与所述第二栅格对应的第一栅格的第一属性信息之间的第二偏差，所述第二栅格的第一属性信息为所述第 二栅格的属性信息中的任意一个属性信息，所述第一栅格的第一属性信息与所述第二栅格的第一属性信息的属性相同；
    根据所述第二偏差，获取所述第一旋转角的第一属性信息对应的第二匹配概率。
  14. 根据权利要求12或13所述的方法，其特征在于，所述根据所述第二匹配概率确定待定位装置的朝向角，包括：
    根据每个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息；
    对所述每种属性信息对应的朝向角信息加权求和，得到所述待定位装置的朝向角。
  15. 根据权利要求1-14中任一项所述的方法，其特征在于，所述获取第一环境对应的栅格地图，包括：
    获取所述第一环境对应的点云的信息，所述第一环境对应的点云的信息用于指示所述第一环境对应的点云中点的位置信息；
    根据所述第一环境对应的点云的信息，获取所述第一环境对应的栅格地图。
  16. 根据权利要求15所述的方法，其特征在于，所述第一环境对应的点云的信息还用于指示所述第一环境对应的点云中点的反射强度。
  17. 根据权利要求1-16中任一项所述的方法，其特征在于，所述获取第二环境对应的局部栅格，包括：
    获取所述第二环境对应的点云的信息，所述第二环境对应的点云的信息用于指示所述第二环境对应的点云中点的位置信息；
    根据所述第二环境对应的点云的信息，获取所述第二环境对应的局部栅格。
  18. 根据权利要求17所述的方法，其特征在于，所述第二环境对应的点云的信息还用于指示所述第二环境对应的点云中点的反射强度。
  19. 一种定位装置，其特征在于，所述装置包括：获取模块和处理模块；
    所述获取模块，用于获取第一环境对应的栅格地图，所述栅格地图包括多个第一栅格，所述第一栅格包括多个点、所述多个点的位置信息和所述第一栅格的属性信息，所述第一栅格的属性信息包括所述第一栅格的语义信息和所述第一栅格的高度信息，所述第一栅格的语义信息用于指示所述第一栅格表示多种物体中的每种物体的概率，所述第一栅格的高度信息用于指示所述第一栅格包括的点的高度；
    所述获取模块，还用于获取第二环境对应的局部栅格，所述第二环境包括在所述第一环境中，所述局部栅格包括至少一个第二栅格，所述第二栅格包括多个点和所述第二栅格的属性信息，所述第二栅格的属性信息包括所述第二栅格的语义信息和所述第二栅格的高度信息，所述第二栅格的语义信息用于指示所述第二栅格表示多种物体中的每种物体的概率，所述第二栅格的高度信息用于指示所述第二栅格包括的点的高度；
    所述获取模块，还用于获取第一匹配概率，所述第一匹配概率为所述局部栅格位于所述栅格地图中的不同位置时，所述局部栅格与所述栅格地图中的栅格的匹配概率；
    所述处理模块，用于根据所述第一匹配概率确定待定位装置的位置，所述待定位装置位于所述第二环境中。
  20. 根据权利要求19所述的定位装置，其特征在于，
    所述第一栅格的属性信息还包括所述第一栅格的反射强度信息，所述第一栅格的反射强度信息用于指示所述第一栅格包括的点的反射强度；
    所述第二栅格的属性信息还包括所述第二栅格的反射强度信息，所述第二栅格的反射强度信息用于指示所述第二栅格包括的点的反射强度。
  21. 根据权利要求19或20所述的定位装置，其特征在于，
    所述处理模块，还用于根据所述待定位装置在第一时刻的位姿信息和所述待定位装置的运动参数，预测所述待定位装置在第二时刻的位置对应的第一区域。
  22. 根据权利要求21所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述局部栅格位于所述第一区域中的栅格时，所述局部栅格与所述栅格地图中的栅格的第一匹配概率。
  23. 根据权利要求22所述的定位装置，其特征在于，
    所述获取模块，具体用于根据所述局部栅格中的栅格的属性信息和所述第一区域中的栅格的属性信息，获取第一区域中第一栅格的每种属性信息对应的第一匹配概率。
  24. 根据权利要求23所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述局部栅格位于所述第一区域中的第一栅格的情况下，所述局部栅格中的第二栅格的第一属性信息与所述第二栅格对应的第一栅格的第一属性信息之间的第一偏差，所述第二栅格的第一属性信息为所述第二栅格的属性信息中的任意一个属性信息，所述第一栅格的第一属性信息与所述第二栅格的第一属性信息的属性相同；
    所述获取模块，还具体用于根据所述第一偏差，获取所述第一区域中的第一栅格的第一属性信息对应的第一匹配概率。
  25. 根据权利要求23或24所述的定位装置，其特征在于，
    所述处理模块，具体用于根据所述第一区域中第一栅格的每种属性信息对应的第一匹配概率，确定每种属性信息对应的位置信息；
    所述处理模块，还具体用于根据所述每种属性信息对应的位置信息，确定所述待定位装置的位置。
  26. 根据权利要求25所述的定位装置，其特征在于，
    所述处理模块，还具体用于在所述每种属性信息对应的位置信息中确定符合预设规则的至少一种属性信息对应的位置信息；
    所述处理模块，还具体用于若所述至少一种属性信息的数量大于1，则对所述至少一种属性信息对应的位置信息加权求和，得到所述待定位装置的位置；或者，若所述至少一种属性信息的数量等于1，则将所述至少一种属性信息对应的位置信息确定为所述待定位装置的位置。
  27. 根据权利要求19-26中任一项所述的定位装置，其特征在于，
    所述获取模块，还用于获取第二匹配概率，所述第二匹配概率为在所述局部栅格位于所述待定位装置的位置，并且所述局部栅格旋转不同角度的情况下，所述局部栅格与所述栅格地图中的栅格的匹配概率；
    所述处理模块，还用于根据所述第二匹配概率确定所述待定位装置的朝向角。
  28. 根据权利要求27所述的定位装置，其特征在于，
    所述处理模块，还用于根据所述待定位装置在第三时刻的朝向角和所述待定位装置的运动参数，预测所述待定位装置在第二时刻的朝向角。
  29. 根据权利要求28所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述局部栅格位于所述待定位装置的位置的情况下，所述局部栅格对应的多个旋转角的第二匹配概率，所述多个旋转角中，每个旋转角与所述预测所述待定位装置在第二时刻的朝向角之间的夹角小于或等于第一角度。
  30. 根据权利要求29所述的定位装置，其特征在于，
    所述获取模块，具体用于根据所述第一旋转角对应的局部栅格中的栅格的属性信息和所述栅格地图中的栅格的属性信息，获取所述第一旋转角的每种属性信息对应的第二匹配概率，所述第一旋转角为所述多个旋转角中的任意一个旋转角。
  31. 根据权利要求30所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述局部栅格位于所述待定位装置的位置，并且所述局部栅格旋转第一旋转角的情况下，所述局部栅格中的每个第二栅格的第一属性信息，与所述第二栅格对应的第一栅格的第一属性信息之间的第二偏差，所述第二栅格的第一属性信息为所述第二栅格的属性信息中的任意一个属性信息，所述第一栅格的第一属性信息与所述第二栅格的第一属性信息的属性相同；
    所述获取模块，还具体用于根据所述第二偏差，获取所述第一旋转角的第一属性信息对应的第二匹配概率。
  32. 根据权利要求30或31所述的定位装置，其特征在于，
    所述处理模块，具体用于根据每个旋转角的每种属性信息对应的第二匹配概率，确定每种属性信息对应的朝向角信息；
    所述处理模块，还具体用于对所述每种属性信息对应的朝向角信息加权求和，得到所述待定位装置的朝向角。
  33. 根据权利要求19-32中任一项所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述第一环境对应的点云的信息，所述第一环境对应的点云的信息用于指示所述第一环境对应的点云中点的位置信息；
    所述获取模块，具体用于根据所述第一环境对应的点云的信息，获取所述第一环境对应的栅格地图。
  34. 根据权利要求33所述的定位装置，其特征在于，所述第一环境对应的点云的信息还用于指示所述第一环境对应的点云中点的反射强度。
  35. 根据权利要求19-34中任一项所述的定位装置，其特征在于，
    所述获取模块，具体用于获取所述第二环境对应的点云的信息，所述第二环境对应的点云的信息用于指示所述第二环境对应的点云中点的位置信息；
    所述获取模块，还具体用于根据所述第二环境对应的点云的信息，获取所述第二环境对应的局部栅格。
  36. 根据权利要求35所述的定位装置，其特征在于，所述第二环境对应的点云的信息还用于指示所述第二环境对应的点云中点的反射强度。
  37. 一种定位装置，其特征在于，包括：处理器，所述处理器与存储器耦合，所述存储器用于存储程序或指令，当所述程序或指令被所述处理器执行时，使得所述装 置执行如权利要求1至18中任一项所述的方法。
  38. 一种计算机可读存储介质，其上存储有计算机程序或指令，其特征在于，所述计算机程序或指令被执行时使得计算机执行如权利要求1至18中任一项所述的方法。
PCT/CN2021/124206 2021-10-15 2021-10-15 定位方法及装置 Ceased WO2023060592A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP21960322.2A EP4404018A4 (en) 2021-10-15 2021-10-15 POSITIONING METHOD AND DEVICE
CN202180103257.8A CN118119907A (zh) 2021-10-15 2021-10-15 定位方法及装置
PCT/CN2021/124206 WO2023060592A1 (zh) 2021-10-15 2021-10-15 定位方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/124206 WO2023060592A1 (zh) 2021-10-15 2021-10-15 定位方法及装置

Publications (1)

Publication Number Publication Date
WO2023060592A1 true WO2023060592A1 (zh) 2023-04-20

Family

ID=85987245

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/124206 Ceased WO2023060592A1 (zh) 2021-10-15 2021-10-15 定位方法及装置

Country Status (3)

Country Link
EP (1) EP4404018A4 (zh)
CN (1) CN118119907A (zh)
WO (1) WO2023060592A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2026048529A (ja) * 2024-09-05 2026-03-17 京セラ株式会社 電子機器、電子機器の制御方法、及びプログラム

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109781119A (zh) * 2017-11-15 2019-05-21 百度在线网络技术（北京）有限公司 一种激光点云定位方法和系统
US20190232487A1 (en) * 2018-02-01 2019-08-01 Robert Bosch Gmbh Method for tracking multiple target objects, device, and computer program for implementing the tracking of multiple target objects for the case of moving objects
CN110221616A (zh) * 2019-06-25 2019-09-10 清华大学苏州汽车研究院（吴江） 一种地图生成的方法、装置、设备及介质
CN110967011A (zh) * 2019-12-25 2020-04-07 苏州智加科技有限公司 一种定位方法、装置、设备及存储介质
CN112836698A (zh) * 2020-12-31 2021-05-25 北京纵目安驰智能科技有限公司 一种定位方法、装置、存储介质及电子设备
CN112867977A (zh) * 2021-01-13 2021-05-28 华为技术有限公司 一种定位方法、装置和车辆
CN113011517A (zh) * 2021-03-30 2021-06-22 上海商汤临港智能科技有限公司 定位结果检测方法、装置、电子设备及存储介质

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111174777A (zh) * 2018-11-09 2020-05-19 阿里巴巴集团控股有限公司 定位方法、装置以及电子设备

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109781119A (zh) * 2017-11-15 2019-05-21 百度在线网络技术（北京）有限公司 一种激光点云定位方法和系统
US20190232487A1 (en) * 2018-02-01 2019-08-01 Robert Bosch Gmbh Method for tracking multiple target objects, device, and computer program for implementing the tracking of multiple target objects for the case of moving objects
CN110221616A (zh) * 2019-06-25 2019-09-10 清华大学苏州汽车研究院（吴江） 一种地图生成的方法、装置、设备及介质
CN110967011A (zh) * 2019-12-25 2020-04-07 苏州智加科技有限公司 一种定位方法、装置、设备及存储介质
CN112836698A (zh) * 2020-12-31 2021-05-25 北京纵目安驰智能科技有限公司 一种定位方法、装置、存储介质及电子设备
CN112867977A (zh) * 2021-01-13 2021-05-28 华为技术有限公司 一种定位方法、装置和车辆
CN113011517A (zh) * 2021-03-30 2021-06-22 上海商汤临港智能科技有限公司 定位结果检测方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
EP4404018A1 (en) 2024-07-24
EP4404018A4 (en) 2024-11-27
CN118119907A (zh) 2024-05-31

Similar Documents

Publication Publication Date Title
US11530924B2 (en) Apparatus and method for updating high definition map for autonomous driving
AU2024219514A1 (en) Computer vision systems and methods for detecting and modeling features of structures in images
CN113587933A (zh) 一种基于分支定界算法的室内移动机器人定位方法
CN114485698B (zh) 一种交叉路口引导线生成方法及系统
Liu et al. Real-time 6d lidar slam in large scale natural terrains for ugv
CN113310488A (zh) 一种基于slam的果园机器人导航方法
US11004332B2 (en) Cooperative mapping for autonomous vehicles, robots or multi-agent systems
CN113838203B (zh) 基于三维点云地图和二维栅格地图的导航系统及应用方法
US20240427019A1 (en) Visual mapping method, and computer program recorded on recording medium for executing method therefor
US20240426987A1 (en) Method for calibration of multiple lidars, and computer program recorded on record-medium for executing method therefor
US20250224252A1 (en) System and method for generating multi-resolution voxel spaces
CN110031825B (zh) 激光定位初始化方法
CN120993920B (zh) 一种复杂场景中的轮式机器人可通过性预测方法及系统
Rachman 3d-lidar multi object tracking for autonomous driving
Wang et al. Map-enhanced ego-lane detection in the missing feature scenarios
WO2024230403A1 (zh) Agv位姿确定方法和装置、及存储介质
CN115060268B (zh) 一种机房融合定位方法、系统、设备及存储介质
CN114966736A (zh) 一种基于点云数据进行目标速度预测的处理方法
Wu et al. A non-rigid hierarchical discrete grid structure and its application to UAVs conflict detection and path planning
CN116858269A (zh) 基于激光slam的烟草行业成品库平库盘点机器人路径优化方法
CN119478465A (zh) 位姿估计方法、计算机设备及存储介质
CN114049362A (zh) 一种基于transformer的点云实例分割方法
WO2024036984A1 (zh) 目标定位方法及相关系统、存储介质
EP4404018A1 (en) Positioning method and device
CN116311114A (zh) 一种可行驶区域生成方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21960322

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202180103257.8

Country of ref document: CN

WWE Wipo information: entry into national phase

Ref document number: 2021960322

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2021960322

Country of ref document: EP

Effective date: 20240419

NENP Non-entry into the national phase

Ref country code: DE