WO2022095023A1 - 一种交通流信息的确定方法、装置、电子设备和存储介质 - Google Patents

一种交通流信息的确定方法、装置、电子设备和存储介质 Download PDF

Info

Publication number
WO2022095023A1
WO2022095023A1 PCT/CN2020/127499 CN2020127499W WO2022095023A1 WO 2022095023 A1 WO2022095023 A1 WO 2022095023A1 CN 2020127499 W CN2020127499 W CN 2020127499W WO 2022095023 A1 WO2022095023 A1 WO 2022095023A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
traffic flow
information
fitting
flow information
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/CN2020/127499
Other languages
English (en)
French (fr)
Inventor
韩佐悦
王子涵
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.)
Uisee Shanghai Automotive Technologies Ltd
Original Assignee
Uisee Shanghai Automotive Technologies 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 Uisee Shanghai Automotive Technologies Ltd filed Critical Uisee Shanghai Automotive Technologies Ltd
Priority to CN202080002861.7A priority Critical patent/CN112567439B/zh
Priority to JP2023527374A priority patent/JP7580155B2/ja
Priority to US18/035,936 priority patent/US12469384B2/en
Priority to KR1020237018513A priority patent/KR102620325B1/ko
Priority to PCT/CN2020/127499 priority patent/WO2022095023A1/zh
Priority to EP20960481.8A priority patent/EP4242998B1/en
Publication of WO2022095023A1 publication Critical patent/WO2022095023A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/161—Decentralised systems, e.g. inter-vehicle communication
    • G08G1/163—Decentralised systems, e.g. inter-vehicle communication involving continuous checking
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/10—Path keeping
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/10—Path keeping
    • B60W30/12—Lane keeping
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/056—Detecting movement of traffic to be counted or controlled with provision for distinguishing direction of travel
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
    • G08G1/096725—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information generates an automatic action on the vehicle control
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems
    • G08G1/167—Driving aids for lane monitoring, lane changing, e.g. blind spot detection
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0019—Control system elements or transfer functions
    • B60W2050/0022—Gains, weighting coefficients or weighting functions
    • B60W2050/0025—Transfer function weighting factor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/10—Number of lanes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2552/00—Input parameters relating to infrastructure
    • B60W2552/53—Road markings, e.g. lane marker or crosswalk
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4041—Position
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4042—Longitudinal speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/404—Characteristics
    • B60W2554/4043—Lateral speed
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W2554/00—Input parameters relating to objects
    • B60W2554/40—Dynamic objects, e.g. animals, windblown objects
    • B60W2554/408—Traffic behavior, e.g. swarm

Definitions

  • the embodiments of the present disclosure relate to the technical field of intelligent driving, and in particular, to a method, an apparatus, an electronic device, and a storage medium for determining traffic flow information.
  • the current intelligent driving system for structured roads in the process of perception of the driving environment, usually locates the lateral position of the vehicle and generates the target driving path based on the lane line information collected by the vehicle's forward-facing camera. It can be seen that the lateral positioning of the vehicle (that is, the positioning of the lane where the vehicle is located), the lateral control of the vehicle and the accuracy of the generation of the target driving path depend on the recognition accuracy of the lane line by the intelligent driving system or the camera.
  • the traffic flow can be understood as the traffic flow formed by multiple vehicles moving in the same direction in the same lane, and the traffic flow information can be understood as the trajectory of the traffic flow. Information.
  • the present disclosure provides a new sensing means capable of sensing traffic flow information. At least one embodiment of the present disclosure provides a method, apparatus, electronic device, and non-transitory computer-readable storage medium for determining traffic flow information.
  • the method for determining traffic flow information proposed by the embodiments of the present disclosure includes:
  • one or more pieces of current traffic flow information are generated by fitting.
  • the grouping of each of the target vehicles based on the motion information of the vehicle and the state information of one or more target vehicles, and obtaining the grouping information includes:
  • each valid target vehicle is grouped to obtain grouping information.
  • the method further includes:
  • a fitting weight of each of the valid target vehicles is determined based on the historical traffic flow information, the state information of each of the valid target vehicles, and the revised grouping information.
  • the grouping of each valid target vehicle based on the driving track of the own vehicle and the status information of each valid target vehicle, and obtaining the grouping information includes:
  • grouping information of the valid target vehicle is determined.
  • the grouping information includes the following five groups: the first lane grouping on the left side of the lane where the vehicle is located, the second lane grouping on the left side of the lane where the vehicle is located, and the first lane grouping on the right side of the lane where the vehicle is located , the second lane grouping on the right side of the lane where the vehicle is located, and other groups.
  • the method before the fitting generates one or more pieces of current traffic flow information, the method further includes:
  • determining a fitting weight of each of the virtual vehicles based on historical traffic flow information, motion information of the host vehicle, state information of each of the virtual vehicles, and grouping information of each of the virtual vehicles;
  • the fitting to generate one or more pieces of current traffic flow information includes: based on the state information of each of the target vehicles, the fitting weight of each of the target vehicles, the state information of each of the virtual vehicles, and The fitting weight of each virtual vehicle is fitted to generate one or more pieces of current traffic flow information.
  • the historical vehicle satisfies a cache condition; wherein, the cache condition is used to filter a vehicle traveling in the same direction as the vehicle and the vehicle does not belong to other groups.
  • the caching conditions include:
  • the vehicle is located in front of the vehicle and the relative distance from the vehicle is greater than the preset relative distance threshold;
  • the angle between the direction of the speed of the vehicle relative to the vehicle and the driving direction of the vehicle is smaller than the preset angle threshold
  • the lateral distance of the vehicle relative to the own vehicle is within a preset lateral distance range.
  • the caching condition further includes: the life cycle of the vehicle is greater than a preset life cycle threshold.
  • the determining a fitting weight for each of the target vehicles includes:
  • the lateral speed deviation weight of the target vehicle Based on the state information of the target vehicle and the grouping information of the target vehicle, determine the lateral speed deviation weight of the target vehicle, the lateral displacement deviation weight of the target vehicle, the life cycle weight of the target vehicle and the speed weight of the target vehicle ;
  • the fitting weight of the target vehicle is obtained by multiplying the lateral speed deviation weight of the target vehicle, the lateral displacement deviation weight of the target vehicle, the life cycle weight of the target vehicle and the speed weight of the target vehicle.
  • the determining a fit weight for each of the virtual vehicles includes:
  • the lateral speed deviation weight of the virtual vehicle Based on the state information of the virtual vehicle and the grouping information of the virtual vehicle, determine the lateral speed deviation weight of the virtual vehicle, the lateral displacement deviation weight of the virtual vehicle, the life cycle weight of the virtual vehicle, and the speed weight of the virtual vehicle and the weight of the virtual vehicle;
  • each group corresponds to a traffic flow; the fitting to generate one or more pieces of current traffic flow information includes:
  • the current traffic flow information corresponding to the group is generated by fitting .
  • selecting the fitting method based on the number of target vehicles in the group and the longitudinal distribution distances of all target vehicles in the group includes:
  • first-order fitting and second-order fitting are selected; otherwise, it is judged whether the number of target vehicles in the group meets the second quantity condition, and the longitudinal distribution distance Whether it is greater than the preset second distance threshold;
  • first-order fitting is selected; otherwise, no fitting is performed.
  • the fitting to generate the current traffic flow information corresponding to the group includes:
  • the current traffic flow corresponding to the group is generated by fitting using the motion information of the own vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group.
  • Information includes:
  • the initial traffic flow information corresponding to the group is generated by fitting
  • the initial traffic flow information is constrained to obtain the current traffic flow information corresponding to the group.
  • the constraints include position constraints, heading constraints, and/or curvature constraints.
  • the location constraint includes: constraining the variation of the constant term in the fitting parameter corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information;
  • the heading constraint includes: constraining a primary term in the fitting parameters corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information and/or the motion information of the host vehicle the amount of change;
  • the curvature constraint includes: constraining the quadratic in the fitting parameters corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information and/or the motion information of the host vehicle The amount of change in the item;
  • the method further includes:
  • the determining a confidence increment based on the current traffic flow information includes:
  • the method further includes: determining a driving reference path of the vehicle based on the current traffic flow information and a confidence level corresponding to the current traffic flow information;
  • the determining of the driving reference path of the vehicle includes:
  • the parameters of the vehicle's driving reference path are determined as the weighted average of the fitting parameters corresponding to the current traffic flow information on both sides of the vehicle. value; wherein, the weight of the fitting parameter corresponding to the current traffic flow information on both sides of the vehicle is inversely correlated with the rate of change of the fitting parameter;
  • the confidence corresponding to the current traffic flow information on only one side of the vehicle is higher than the preset confidence threshold, determine the fitting of the primary term and the secondary term in the parameters of the vehicle's reference path and the current traffic flow information on that side.
  • the primary and secondary terms in the parameters are the same, and the constant term in the parameters of the reference path of the vehicle is zero;
  • the parameters for determining the reference path for the vehicle to travel are all zero.
  • the method further includes: determining an auxiliary positioning marker based on the current traffic flow information and a confidence level corresponding to the current traffic flow information;
  • the determining of the auxiliary positioning flag includes:
  • the auxiliary positioning mark is determined as the first mark
  • the auxiliary positioning mark is determined as the second mark
  • the auxiliary positioning mark is determined as the third mark.
  • the apparatus for determining traffic flow information includes:
  • a grouping module configured to group each of the target vehicles based on the motion information of the vehicle and the state information of one or more target vehicles to obtain grouping information
  • a determining module configured to determine the fitting weight of each of the target vehicles based on the state information of each of the target vehicles and the grouping information
  • a fitting module configured to fit and generate one or more pieces of current traffic based on the motion information of the vehicle, the state information of each of the target vehicles, the fitting weight of each of the target vehicles and the grouping information flow information.
  • an electronic device provided by an embodiment of the present disclosure includes: a processor and a memory; the processor is configured to execute the traffic flow information according to any embodiment of the first aspect by calling a program or an instruction stored in the memory the steps of the determination method.
  • a non-transitory computer-readable storage medium provided by an embodiment of the present disclosure is used to store a program or an instruction, and the program or instruction causes a computer to execute the method for determining traffic flow information according to any embodiment of the first aspect A step of.
  • FIG. 1 is an exemplary scene diagram for determining traffic flow information provided by an embodiment of the present disclosure
  • FIG. 2 is an exemplary architecture diagram of an intelligent driving vehicle provided by an embodiment of the present disclosure
  • FIG. 3 is an exemplary block diagram of an electronic device provided by an embodiment of the present disclosure.
  • FIG. 4 is an exemplary flowchart of a method for determining traffic flow information provided by an embodiment of the present disclosure
  • FIG. 5 is an exemplary block diagram of an apparatus for determining traffic flow information provided by an embodiment of the present disclosure.
  • the embodiments of the present disclosure provide a new perception method capable of perceiving traffic flow information.
  • the traffic flow can be understood as the traffic flow formed by multiple vehicles moving in the same direction in the same lane, and the traffic flow information can be understood as the information of the trajectory of the traffic flow. . Since the lane line can be regarded as a line, and the trajectory can also be regarded as a line, the traffic flow information can replace the lane line for vehicle lateral positioning (that is, the positioning of the lane where the vehicle is located), vehicle lateral control and targeting Driving path generation.
  • the embodiments of the present disclosure provide a method, device, electronic device, or storage medium for determining traffic flow information. By grouping multiple target vehicles and determining the fitting weight of each target vehicle, the fitting weight of each target vehicle can be determined. The combined weight, the motion information of the vehicle and the state information of the target vehicle are fitted to generate the current traffic flow information to realize the perception of the traffic flow.
  • the embodiments of the present disclosure can be applied to intelligent driving vehicles, and can also be applied to electronic devices.
  • the intelligent driving vehicle is a vehicle equipped with different levels of intelligent driving systems.
  • the intelligent driving systems include, for example, an unmanned driving system, an assisted driving system, a driving assistance system, a highly automatic driving system, a fully automatic driving vehicle, and the like.
  • the electronic device is installed with an intelligent driving system.
  • the electronic device can be used to test the intelligent driving algorithm.
  • the electronic device can be a vehicle-mounted device.
  • the electronic device can also be applied to other fields.
  • the embodiments of the present disclosure take an intelligent driving vehicle as an example to describe the method, apparatus, electronic device, or storage medium for determining the traffic flow information.
  • Fig. 1 is an exemplary scene diagram for determining traffic flow information provided by an embodiment of the present disclosure.
  • the vehicle 101 is going straight, the left and right boundaries of the vehicle's trajectory are shown as 104, and there are 3 vehicles on the left side of the vehicle
  • the traffic flow information on the left side of the vehicle can be determined by the method for determining traffic flow information provided by the embodiment of the present disclosure, for example, the traffic flow on the left side in FIG. 1
  • the corresponding trajectory 102, and the traffic flow information on the right side of the vehicle is determined.
  • FIG. 2 is an exemplary overall architecture diagram of an intelligent driving vehicle according to an embodiment of the present disclosure.
  • the intelligent driving vehicle shown in FIG. 2 may be implemented as the vehicle 101 in FIG. 1 .
  • an intelligent driving vehicle includes: a sensor group, an intelligent driving system, a vehicle underlying execution system, and other components that can be used to drive the vehicle and control the operation of the vehicle, such as the brake pedal, steering wheel, and accelerator pedal.
  • the sensor group is used to collect the data of the external environment of the vehicle and detect the position data of the vehicle.
  • the sensor group includes, but is not limited to, at least one of a camera, a lidar, a millimeter-wave radar, an ultrasonic radar, a GPS (Global Positioning System, global positioning system), and an IMU (Inertial Measurement Unit, inertial measurement unit).
  • the sensor group is also used to collect dynamic data of the vehicle, for example, the sensor group also includes but is not limited to at least one of a wheel speed sensor, a speed sensor, an acceleration sensor, a steering wheel angle sensor, and a front wheel angle sensor.
  • the intelligent driving system is used to acquire sensing data of a sensor group, wherein the sensing data includes but is not limited to images, videos, laser point clouds, millimeter waves, GPS information, vehicle status, and the like.
  • the intelligent driving system performs environmental perception and vehicle positioning based on the sensing data, and generates sensing information and vehicle posture; performs planning and decision-making based on the sensing information and vehicle posture, and generates planning and decision-making information; Based on the planning and decision-making information, vehicle control commands are generated and sent to the underlying vehicle execution system.
  • the control instructions may include, but are not limited to, steering wheel steering, lateral control instructions, longitudinal control instructions, and the like.
  • the intelligent driving system acquires sensor data, V2X (Vehicle to X, vehicle wireless communication) data, high-precision maps and other data, and performs environmental perception and positioning based on at least one of the above data, and generates perception information and positioning information.
  • the perception information may include, but is not limited to, at least one of the following: obstacle information, road signs/marks, pedestrian/vehicle information, and drivable areas.
  • the positioning information includes the vehicle pose.
  • the intelligent driving system generates planning and decision-making information based on perception information and vehicle pose, as well as at least one of V2X data, high-precision maps, and other data.
  • planning information may include but not limited to planning paths, etc.
  • decision information may include but not limited to at least one of the following: behavior (for example, including but not limited to following, overtaking, parking, detouring, etc.), vehicle heading, vehicle speed, Desired acceleration of the vehicle, desired steering wheel angle, etc.
  • the method for determining traffic flow information provided by the embodiments of the present disclosure may be applied to the intelligent driving system.
  • the intelligent driving system may be a software system, a hardware system, or a system combining software and hardware.
  • an intelligent driving system is a software system running on an operating system
  • an in-vehicle hardware system is a hardware system that supports the operation of the operating system.
  • the intelligent driving system can interact with a cloud server.
  • the intelligent driving system interacts with the cloud server through a wireless communication network (for example, including but not limited to GPRS network, Zigbee network, Wifi network, 3G network, 4G network, 5G network and other wireless communication networks).
  • a wireless communication network for example, including but not limited to GPRS network, Zigbee network, Wifi network, 3G network, 4G network, 5G network and other wireless communication networks.
  • a cloud server is used to interact with the vehicle.
  • the cloud server can send the environment information, positioning information, control information and other information required in the intelligent driving process of the vehicle to the vehicle.
  • the cloud server may receive sensor data, vehicle status information, vehicle driving information and related information requested by the vehicle from the vehicle end.
  • the cloud server may remotely control the vehicle based on user settings or vehicle request.
  • the cloud server may be a server or a server group. Server farms can be centralized or distributed. In some embodiments, the cloud server may be local or remote.
  • the vehicle bottom execution system is used to receive vehicle control instructions, and control the vehicle to drive based on the vehicle control instructions.
  • the underlying vehicle execution systems include, but are not limited to, steering systems, braking systems, and drive systems.
  • the vehicle bottom-level execution system may further include a bottom-level controller, which can parse the vehicle control instructions and deliver them to corresponding systems such as a steering system, a braking system, and a driving system, respectively.
  • the intelligent driving vehicle may further include a vehicle CAN bus not shown in FIG. 2 , and the vehicle CAN bus is connected to the vehicle bottom execution system. The information interaction between the intelligent driving system and the vehicle bottom execution system is transmitted through the vehicle CAN bus.
  • FIG. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.
  • the electronic device may be an in-vehicle device.
  • the electronic device may support the operation of the intelligent driving system.
  • the electronic device includes: at least one processor 301 , at least one memory 302 and at least one communication interface 303 .
  • the various components in the in-vehicle device are coupled together by a bus system 304 .
  • the communication interface 303 is used for information transmission with external devices. Understandably, the bus system 304 is used to implement connection communication between these components.
  • the bus system 304 also includes a power bus, a control bus and a status signal bus. However, for clarity of illustration, the various buses are labeled as bus system 304 in FIG. 3 .
  • the memory 302 in this embodiment may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory.
  • memory 302 stores the following elements, executable units or data structures, or a subset thereof, or an extended set of them: an operating system and an application program.
  • the operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic tasks and processing hardware-based tasks.
  • Applications including various applications, such as media players (Media Player), browsers (Browser), etc., are used to implement various application tasks.
  • a program for implementing the method for determining traffic flow information provided by the embodiments of the present disclosure may be included in an application program.
  • the processor 301 calls the program or instruction stored in the memory 302, specifically, the program or instruction stored in the application program, and the processor 301 is configured to execute the traffic flow information provided by the embodiment of the present disclosure. The steps of each embodiment of the method are determined.
  • the method for determining traffic flow information may be applied to the processor 301 or implemented by the processor 301 .
  • the processor 301 may be an integrated circuit chip, which has signal processing capability. In the implementation process, each step of the above-mentioned method can be completed by an integrated logic circuit of hardware in the processor 301 or an instruction in the form of software.
  • the above-mentioned processor 301 can be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), a ready-made programmable gate array (Field Programmable Gate Array, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
  • the steps of the method for determining traffic flow information provided by the embodiments of the present disclosure may be directly embodied as executed by a hardware decoding processor, or executed by a combination of hardware and software units in the decoding processor.
  • the software unit may be located in random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers and other storage media mature in the art.
  • the storage medium is located in the memory 302, and the processor 301 reads the information in the memory 302, and completes the steps of the method in combination with its hardware.
  • FIG. 4 is an exemplary flowchart of a method for determining traffic flow information according to an embodiment of the present disclosure.
  • the executing subject of the method is an electronic device.
  • the executing subject of the method may also be an intelligent driving system supported by the electronic device.
  • the intelligent driving system is used as the execution subject to describe the flow of the method for determining the traffic flow information.
  • the intelligent driving system groups each target vehicle based on the motion information of the vehicle and the state information of one or more target vehicles to obtain grouping information.
  • the motion information of the vehicle is dynamic information related to driving during the driving process of the vehicle.
  • the motion information of the vehicle may include, but is not limited to, wheel rotation speed, vehicle speed, acceleration, steering wheel angle, front wheel angle, and yaw rate.
  • the motion information of the vehicle may be collected by the sensor group shown in FIG. 1 , and the intelligent driving system may obtain the motion information of the vehicle from the sensor group.
  • the intelligent driving system obtains data such as sensor data, V2X data, and high-precision maps, and performs environmental perception based on at least one of the above data, and can obtain status information of one or more target vehicles.
  • the state information of the target vehicle is a general term for information such as relative position and relative motion between the target vehicle and the host vehicle.
  • the state information of the target vehicle may include, but is not limited to, at least one of the lateral displacement, longitudinal displacement, lateral velocity, and longitudinal velocity of the target vehicle relative to the host vehicle, wherein the lateral direction can be understood as the lateral direction, that is, the vertical direction in the direction of the lane line.
  • the state information of the target vehicle may also include, but is not limited to, one or more of identification (ID), perception type, life cycle, and target confidence.
  • ID identification
  • the life cycle can be understood as the time period between the time when the target vehicle is first perceived to the current time, that is, the time period when the target vehicle appears in the "field of view" of the vehicle.
  • the target confidence level is used to represent the credibility of the target vehicle and is automatically generated by the perception algorithm of the intelligent driving system.
  • the grouping rule adopted by the intelligent driving system is to divide the grouping of target vehicles into five groups according to the lane where the vehicle is located: the first lane on the left side of the lane where the vehicle is located is grouped , the second lane grouping on the left side of the lane where the vehicle is located, the first lane grouping on the right side of the lane where the vehicle is located, the second lane grouping on the right side of the lane where the vehicle is located, and other groups.
  • other groups are any situation other than the foregoing four groups.
  • other groupings may be one or more of the lane where the vehicle is located, the third lane to the left of the lane where the vehicle is located, and the third lane to the right of the lane where the vehicle is located.
  • the intelligent driving system may mark each target vehicle with a group mark, and the group mark is used to represent the group to which the target vehicle belongs.
  • the group mark may be a group flag bit, Different values of the grouping flag indicate different groups.
  • the grouping information includes at least a grouping flag.
  • the grouping information may also include, but is not limited to, one or more of the target vehicle's ID, perception type, life cycle, and target confidence.
  • the intelligent driving system determines the driving trajectory of the vehicle based on the motion information of the vehicle; Determine the relative lane where each target vehicle is located (the relative lane is relative to the lane where the vehicle is located); thus, according to the aforementioned grouping rules, each target vehicle can be grouped based on the relative lane where each target vehicle is located to obtain a grouping information.
  • the intelligent driving system when determining the driving trajectory of the vehicle, calculates the turning radius of the vehicle based on the signals of the vehicle speed, steering wheel angle and yaw rate after filtering, and then determines the current vehicle based on the turning radius. 's driving trajectory. It should be noted that the determination of the driving trajectory by the turning radius is a mature technology in the field of intelligent driving, and the specific determination process will not be repeated here.
  • the intelligent driving system first filters valid target vehicles and filters invalid target vehicles; A valid target vehicle is grouped to obtain grouping information.
  • the intelligent driving system screens one or more valid target vehicles based on the motion information of the vehicle and the status information of each target vehicle. For example, the intelligent driving system performs screening by removing invalid target vehicles.
  • the invalid target vehicles include: target vehicles whose longitudinal speed is negative (the speed of the own vehicle is positive) or target vehicles whose longitudinal distance is greater than a preset longitudinal distance threshold. That is, the longitudinal speed of the valid target vehicle is positive (or zero) and the longitudinal distance is less than or equal to the preset longitudinal distance threshold.
  • the vehicle may be changing lanes, steering, and other non-straight-traveling motions, which may cause grouping errors.
  • the vehicle is divided into the first lane grouping on the left side of the lane where the vehicle is located.
  • the target vehicle may be divided into the second lane grouping on the left side of the lane where the vehicle is located, resulting in grouping errors. Therefore, it is necessary to correct the grouping information to avoid grouping errors caused by the rapid turning of the vehicle.
  • the intelligent driving system modifies the grouping information based on historical traffic flow information.
  • the historical traffic flow information is generated by fitting in the last definite period, where the definite period is the period for determining the traffic flow information preset by the intelligent driving system. Set the determination period, ranging from tens of milliseconds to several seconds, for example, set the determination period to 5 seconds.
  • Each group corresponds to a traffic flow, that is, a piece of historical traffic flow information corresponds to a group.
  • the historical traffic flow information includes the value of a fitting parameter, and the fitting parameter includes a quadratic term, a linear term and a constant term. If the quadratic term exists, the historical traffic flow information corresponds to a curve.
  • the historical traffic flow information corresponds to a straight line.
  • the historical traffic flow information further includes a confidence level, and the confidence level is used to represent the reliability of the traffic flow. If there is no historical traffic flow information, for example, when the vehicle leaves the garage and enters a low-speed lane, the historical traffic flow information has not been stored in the intelligent driving system, then the confidence level of the historical traffic flow information is zero, and the grouping information is not corrected.
  • the intelligent driving system determines whether there is a target vehicle within a preset range around the historical traffic flow information, and if so, corrects the grouping of the target vehicle to the historical traffic flow information.
  • the grouping corresponding to the traffic flow information.
  • the preset range may be set according to actual needs, and this embodiment does not limit the specific value of the preset range. For example, a target vehicle is grouped into the second lane on the left, and the target vehicle is located within a preset range around the historical traffic flow information, and the historical traffic flow information corresponds to the first lane on the left.
  • the grouping of vehicles is corrected from the second lane grouping on the left to the first lane grouping on the left.
  • the intelligent driving system buffers the vehicles that meet the buffering conditions in each corrected group. For the next determination period, the buffered vehicles are historical vehicles.
  • the cache duration may be set, and the cache duration may be set based on actual needs. This embodiment does not limit the specific value of the cache duration.
  • the cache condition is used to filter vehicles traveling in the same direction as the vehicle and the vehicle does not belong to other groups. It should be noted that, for each determination period, the intelligent driving system will cache the vehicles that meet the caching conditions in the corrected groups in the determination period, that is, the timing of caching is within the determination period.
  • step 402 the intelligent driving system determines the fitting weight of each target vehicle based on the state information and grouping information of each target vehicle.
  • the intelligent driving system for each target vehicle determines the lateral speed of the target vehicle based on the state information of the target vehicle and the grouping information of the target vehicle The deviation weight, the lateral displacement deviation weight of the target vehicle, the life cycle weight of the target vehicle and the speed weight of the target vehicle; and then the lateral speed deviation weight of the target vehicle, the lateral displacement deviation weight of the target vehicle, The life cycle weight of the target vehicle and the speed weight of the target vehicle are multiplied to obtain the fitting weight of the target vehicle.
  • the calculation method of the lateral speed deviation weight of the target vehicle is: calculating the average lateral speed of all target vehicles in the group to which the target vehicle belongs; based on the lateral speed of the target vehicle and the average lateral speed, calculating the The lateral speed deviation weight of the target vehicle. In this embodiment, if the lateral speed of the target vehicle deviates more from the average lateral speed, the weight of the lateral speed deviation is lower.
  • the calculation method of the lateral displacement deviation weight of the target vehicle is as follows: calculate the average lateral displacement of all target vehicles in the group to which the target vehicle belongs; based on the lateral displacement of the target vehicle and the average lateral displacement, calculate the target vehicle The lateral displacement of deviates from the weight. In this embodiment, if the lateral displacement of the target vehicle deviates farther from the average lateral displacement, the lateral displacement deviation weight is lower.
  • the calculation method of the life cycle weight of the target vehicle is: calculating the life cycle weight of the target vehicle based on the life cycle of the target vehicle and/or the target confidence.
  • the longer the life cycle of the target vehicle the higher the life cycle weight.
  • the calculation method of the speed weight of the target vehicle is: if there is no historical traffic flow information corresponding to the grouping of the target vehicle, the speed weight of the target vehicle is 1.
  • the lateral displacement deviation weight of the target vehicle may be determined in conjunction with historical traffic flow information. Specifically, the lateral displacement deviation weight of the target vehicle is calculated based on the lateral displacement of the target vehicle and the historical traffic flow information corresponding to the grouping of the target vehicle. In this embodiment, if the lateral displacement of the target vehicle deviates farther from the historical traffic flow, the lateral displacement deviation weight is lower.
  • the speed weight of the target vehicle may be determined in conjunction with historical traffic flow information. Specifically, based on the lateral speed and longitudinal speed of the target vehicle, the driving speed of the target vehicle is synthesized, and then the driving direction of the target vehicle is determined; thus, based on the driving direction of the target vehicle and the history corresponding to the grouping of the target vehicle Traffic flow information, calculate the speed weight of the target vehicle. In this embodiment, if the included angle between the traveling direction of the target vehicle and the tangential direction of the historical traffic flow is larger, the speed weight of the target vehicle is lower.
  • the intelligent driving system after grouping each valid target vehicle to obtain grouping information, and using historical traffic flow information to correct the grouping information, the intelligent driving system based on the historical traffic flow information, the state information of each valid target vehicle and the correction After grouping information, the fitting weight of each valid target vehicle is determined.
  • step 403 the intelligent driving system generates one or more current traffic flow information by fitting based on the motion information of the vehicle, the state information of each target vehicle, the fitting weight and grouping information of each target vehicle.
  • each group corresponds to a traffic flow
  • the intelligent driving system generates current traffic flow information corresponding to the group by fitting, specifically, based on the number of target vehicles in the group and all targets in the group The longitudinal distribution distance of the vehicle, select the fitting method; and then based on the selected fitting method, use the motion information of the vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group to fit.
  • the current traffic flow information corresponding to the group is synthesized.
  • the traffic flow obtained by grouping fitting of the first lane on the left is called the traffic flow on the left side of the vehicle; the traffic flow obtained by grouping the first lane on the right is called the traffic flow on the right side of the vehicle.
  • the intelligent driving system determines whether the number of target vehicles in the group satisfies the first number condition, and whether the longitudinal distribution distance is greater than a preset first distance threshold; wherein the first number The condition is, for example, 3, and the first distance threshold is, for example, 50 meters, to avoid overfitting (for example, the traffic flow is actually a straight line, but the curve is used for fitting), and the specific values of the first quantity condition and the first distance threshold can be Set according to actual needs.
  • first-order fitting and second-order fitting are selected; otherwise, it is judged whether the number of target vehicles in the group meets the second quantity condition, and the longitudinal distribution distance Whether it is greater than a preset second distance threshold; wherein, the second quantity condition is, for example, greater than or equal to 2 and less than 4, the second distance threshold is, for example, 30 meters, and the specific values of the first quantity condition and the first distance threshold can be Set according to actual needs.
  • first-order fitting is selected; otherwise, the fitting is not performed, and the data of the previous traffic flow information determination period is continued.
  • the fitting when the intelligent driving system generates the current traffic flow information corresponding to the group by fitting, the fitting generates the first-order fitting result corresponding to the group, and determines The first mean square error of the first-order fitting result; the fitting generates the second-order fitting result corresponding to the group, and determines the second mean square error of the second-order fitting result; and then compares the first mean square error and the second mean square error, if If the first mean square error is greater than the preset multiple of the second mean square error (for example, 1.5 times, the preset multiple can also be set according to the actual situation), the second-order fitting result is selected as the current traffic flow information corresponding to the group; otherwise, a The order fitting result is the current traffic flow information corresponding to the group.
  • the preset multiple of the second mean square error for example, 1.5 times, the preset multiple can also be set according to the actual situation
  • the driving system uses the motion information of the vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group, when fitting the current traffic flow information corresponding to the group
  • the driving system uses the motion information of the vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group, when fitting the current traffic flow information corresponding to the group
  • the driving system uses the motion information of the vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group to fitting; and then based on the historical traffic flow information and the motion information of the vehicle , the initial traffic flow information is constrained to obtain the current traffic flow information corresponding to the group.
  • the constraints on the initial traffic flow information include position constraints, heading constraints, and/or curvature constraints.
  • the location constraint includes: constraining the variation of the constant term in the fitting parameter corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information; wherein the variation of the constant term corresponds to the historical traffic flow information
  • the number of fittings can be understood as the number of target vehicles participating in the fitting.
  • the heading constraint includes: based on the confidence level corresponding to the historical traffic flow information and/or the motion information of the vehicle, constraining the variation of the primary item in the fitting parameter corresponding to the initial traffic flow information; wherein the primary item There is an inverse correlation between the variation of the primary item and the confidence level corresponding to the historical traffic flow information, and there is a positive correlation between the variation of the primary item and the steering amplitude of the vehicle, and/or the variation of the primary item corresponds to the initial traffic flow information. There is a positive correlation between the fitted quantities.
  • the curvature constraint includes: based on the confidence corresponding to the historical traffic flow information and/or the motion information of the vehicle, constraining the variation of the quadratic term in the fitting parameters corresponding to the initial traffic flow information; wherein the two There is an inverse correlation between the amount of change in the secondary term and the confidence level corresponding to the historical traffic flow information, there is an inverse correlation between the amount of change in the quadratic term and the speed of the vehicle, and/or the amount of change in the quadratic term is inversely correlated with the speed of the vehicle. There is a positive correlation between the steering wheel speed.
  • each group corresponds to a traffic flow, and for a group, before fitting and generating the current traffic flow information corresponding to the group, for example, after grouping each target vehicle and obtaining the group information, the intelligent driving system Make sure that the number of target vehicles in this group meets the preset number of fittings, otherwise the fitting cannot be performed. Therefore, the intelligent driving system determines whether the number of target vehicles meets the preset fitting number; if so, it determines whether the driving state of the vehicle is straight; if it is straight, it acquires the cached historical vehicles, and the acquired historical vehicles can be One or more, and the acquired historical vehicles belong to this group.
  • the historical vehicle satisfies a cache condition; the cache condition is used to filter a vehicle traveling in the same direction as the vehicle and the vehicle does not belong to other groups.
  • the caching conditions include the following (1) to (3):
  • the vehicle is located in front of the vehicle and the relative distance from the vehicle is greater than the preset relative distance threshold; wherein, the preset relative distance threshold is, for example, 20 meters.
  • the preset relative distance threshold is, for example, 20 meters.
  • This embodiment does not limit the specific value of the relative distance threshold.
  • Those skilled in the art can make settings according to actual needs. It can be understood that when determining whether a vehicle satisfies the cache condition, it is based on the relative relationship between the vehicle and the vehicle at the time of determination, including relative distance, driving direction, lateral distance, and the like.
  • the included angle between the direction of the speed of the vehicle relative to the own vehicle and the driving direction of the own vehicle is smaller than a preset included angle threshold.
  • the direction of the speed of the vehicle relative to the vehicle may be understood as the driving direction of the vehicle.
  • the driving direction of the vehicle may be based on the lateral speed and the longitudinal speed of the vehicle at the judgment moment, and the speed of the vehicle (that is, the driving speed) can be synthesized to determine the driving direction of the vehicle.
  • the lateral distance of the vehicle relative to the own vehicle is within a preset lateral distance range.
  • the preset lateral distance range is, for example, 2 meters to 3 meters. This embodiment does not limit the specific value of the lateral distance range, which can be set by those skilled in the art according to actual needs.
  • the cache condition may also include the following (4):
  • the life cycle of the vehicle is greater than the preset life cycle threshold.
  • the preset life cycle threshold is, for example, 3 seconds. This embodiment does not limit the specific value of the life cycle threshold, and those skilled in the art can set it according to actual needs.
  • the intelligent driving system updates the coordinates of each historical vehicle to obtain the virtual vehicle corresponding to each historical vehicle and the status information of each virtual vehicle.
  • the way of updating the coordinates is, for example, calculating the time difference between the cached time of the historical vehicle and the current time, using the state information of the historical vehicle and the calculated time difference to update the coordinates, and specifically, using the lateral speed and longitudinal direction in the state information of the historical vehicle. Speed, synthesizing the traveling speed of the historical vehicle, and updating the coordinates based on the traveling speed and the calculated time difference.
  • the state information of the historical vehicle is the state information corresponding to the historical vehicle when it is cached.
  • the state information of the virtual vehicle includes, but is not limited to, at least one of lateral displacement, longitudinal displacement, lateral velocity, and longitudinal velocity of the virtual vehicle relative to the host vehicle.
  • the lateral displacement and the longitudinal displacement are determined based on the aforementioned coordinate update, and the lateral speed and the longitudinal speed are the same as the lateral and longitudinal speeds of the corresponding historical vehicle.
  • the status information of the virtual vehicle may also include, but is not limited to, one or more of identification (ID), perception type, life cycle, and target confidence, etc., which information is the same as the historical vehicle corresponding to the virtual vehicle , for example, the perception type of the virtual vehicle is the perception type of the corresponding historical vehicle.
  • the grouping information of the virtual vehicle is the same as the grouping information of the corresponding historical vehicle.
  • the historical vehicle is not screened, because the relative relationship between the historical vehicle and the vehicle cannot be determined, and the coordinates of the historical vehicle cannot be updated to obtain the virtual vehicle, even if the historical vehicle The virtual vehicle is obtained by updating the coordinates of , but since the relative relationship between the historical vehicle and the own vehicle is uncertain, the relative relationship between the virtual vehicle and the own vehicle is also uncertain, so the state information of the virtual vehicle cannot be determined.
  • the intelligent driving system may, based on the historical traffic flow information, the motion information of the vehicle, the state information of each virtual vehicle, and the grouping information of each virtual vehicle, Determine the fitting weights for each virtual vehicle.
  • the intelligent driving system when determining the fitting weight of each virtual vehicle, determines the lateral speed of the virtual vehicle based on the state information of the virtual vehicle and the grouping information of the virtual vehicle The deviation weight, the lateral displacement deviation weight of the virtual vehicle, the life cycle weight of the virtual vehicle, the speed weight of the virtual vehicle, and the weight of the virtual vehicle; and then the lateral speed deviation weight of the virtual vehicle, the virtual vehicle's The lateral displacement deviation weight, the life cycle weight of the virtual vehicle, the speed weight of the virtual vehicle and the weight of the virtual vehicle are multiplied to obtain the fitting weight of the virtual vehicle.
  • the calculation methods of the lateral speed deviation weight of the virtual vehicle, the lateral displacement deviation weight of the virtual vehicle, the life cycle weight of the virtual vehicle, and the speed weight of the virtual vehicle are respectively the same as the lateral speed deviation weight of the aforementioned target vehicle.
  • the lateral displacement deviation weight of the target vehicle, the life cycle weight of the target vehicle, and the speed weight of the target vehicle are calculated in the same manner, and are not repeated here.
  • the weight of the virtual vehicle is calculated based on the cache duration of the virtual vehicle. The longer the cache time of the virtual vehicle, the lower the weight of the virtual vehicle.
  • the intelligent driving system is based on the state information of each target vehicle, the fitting weight of each target vehicle, the state information of each virtual vehicle, and the state information of each virtual vehicle.
  • the fitting weight of the vehicle the fitting generates one or more pieces of current traffic flow information.
  • the process of fitting generation is similar to the above-mentioned process of fitting only based on the target vehicle, and details are not repeated here.
  • the intelligent driving system may determine the confidence increment based on the current traffic flow information; and then determine the current traffic flow based on the confidence and confidence increment of the historical traffic flow information The confidence level corresponding to the flow information. For example, the confidence level based on the historical traffic flow information is added to the confidence level increment to obtain the confidence level corresponding to the current traffic flow information. It should be noted that only when the current traffic flow information and the historical traffic flow information correspond to the same group, the confidence level corresponding to the current traffic flow information can be determined by the method of this embodiment.
  • the intelligent driving system determines whether there is an intersection of two current traffic flow information, and if so, compares the fitting quantity and longitudinal distribution distance corresponding to the two traffic flows; The confidence increment corresponding to the current traffic flow information with a small number or a short longitudinal distribution distance is negative. For example, the intelligent driving system determines whether the traffic flow on the left side of the vehicle and the traffic flow on the right side of the vehicle intersect. The confidence increment corresponding to the current traffic flow information with a short distribution distance is negative.
  • the confidence increment is negative. This negative value will be accumulated to the confidence of the corresponding historical traffic flow information.
  • the intelligent driving system may determine the driving reference path of the vehicle based on the current traffic flow information and the confidence level corresponding to the current traffic flow information. Determining the driving reference path of the vehicle includes:
  • the parameters of the vehicle's driving reference path are determined as the weighted average of the fitting parameters corresponding to the current traffic flow information on both sides of the vehicle. value; wherein, the weight of the fitting parameter corresponding to the current traffic flow information on both sides of the vehicle is inversely correlated with the rate of change of the fitting parameter;
  • the confidence corresponding to the current traffic flow information on only one side of the vehicle is higher than the preset confidence threshold, determine the fitting of the primary term and the secondary term in the parameters of the vehicle's reference path and the current traffic flow information on that side.
  • the primary and secondary terms in the parameters are the same, and the constant term in the parameters of the reference path of the vehicle is zero;
  • the parameters for determining the reference path for the vehicle to travel are all zero.
  • the intelligent driving system may determine the auxiliary positioning marker based on the current traffic flow information and the confidence level corresponding to the current traffic flow information.
  • the auxiliary positioning mark is used to assist in determining the lane where the vehicle is located.
  • the auxiliary positioning sign is independent of the lane line information. In some multi-lane scenarios, when the lane line is not enough to determine the only lane, it can assist in determining the lane where the vehicle is located, providing a necessary basis for the long-distance path planning of the L3-level automatic driving system.
  • determining an auxiliary positioning marker includes:
  • the auxiliary positioning mark is determined as the first mark.
  • the first sign indicates that there are lanes on both sides of the vehicle.
  • the auxiliary positioning flag may be a flag bit, and the value of the flag bit corresponding to the first flag is, for example, 1. When the intelligent driving system determines that the auxiliary positioning flag is 1, it is determined that there are lanes on both sides of the vehicle.
  • the auxiliary positioning mark is determined as the second mark.
  • the second sign indicates that there is a lane on the left side of the vehicle.
  • the value of the flag bit corresponding to the second flag is, for example, 2.
  • the auxiliary positioning mark is determined as the third mark.
  • the third sign indicates that there is a lane on the right side of the vehicle.
  • the value of the flag bit corresponding to the third flag is, for example, 3.
  • the intelligent driving system can perform lateral control of the vehicle based on the current traffic flow information, which can get rid of the dependence on the lane line information, and is conducive to maintaining the lateral control of the vehicle in a congested environment.
  • Performing lateral control based on a path is a mature technical means in the art, and details are not described herein again.
  • Embodiments of the present disclosure further provide a non-transitory computer-readable storage medium, where the non-transitory computer-readable storage medium stores programs or instructions, the programs or instructions cause a computer to execute various embodiments of the method for determining traffic flow information, for example In order to avoid repeated description, the steps are not repeated here.
  • FIG. 5 is an exemplary block diagram of an apparatus for determining traffic flow information provided by an embodiment of the present disclosure.
  • the apparatus for determining traffic flow information includes, but is not limited to, a grouping module 501 , a determining module 502 and a fitting module 503 .
  • the grouping module 501 is configured to group each of the target vehicles based on the motion information of the own vehicle and the state information of one or more target vehicles to obtain grouping information.
  • a determination module 502 configured to determine a fitting weight of each of the target vehicles based on the state information of each of the target vehicles and the grouping information.
  • the fitting module 503 is used for fitting to generate one or more current Traffic flow information.
  • the grouping module 501 is configured to determine the driving track of the own vehicle based on the motion information of the own vehicle; based on the motion information of the own vehicle and the state information of each target vehicle, filter one or more A plurality of valid target vehicles; based on the driving track of the vehicle and the state information of each valid target vehicle, each valid target vehicle is grouped to obtain grouping information.
  • the grouping module 501 is further configured to modify the grouping information based on the historical traffic flow information after the grouping information is obtained; correspondingly, the determining module 502 is configured to, based on the historical traffic flow information, each The state information and the revised grouping information of each of the valid target vehicles are determined, and the fitting weight of each valid target vehicle is determined.
  • the grouping module 501 for each valid target vehicle: determines the relative lane where the valid target vehicle is located based on the state information of the valid target vehicle and the driving track of the own vehicle; Lane, to determine the grouping information of the valid target vehicle.
  • the grouping information includes the following five groups: the first lane grouping on the left side of the lane where the vehicle is located, the second lane grouping on the left side of the lane where the vehicle is located, and the first lane grouping on the right side of the lane where the vehicle is located , the second lane grouping on the right side of the lane where the vehicle is located, and other groups.
  • the fitting module 503 is further configured to judge whether the number of target vehicles satisfies the preset fitting quantity before fitting to generate one or more pieces of current traffic flow information; if so, judge the driving of the vehicle Whether the state is going straight; if the driving state of the vehicle is going straight, obtain the cached historical vehicles; update the coordinates of each historical vehicle to obtain the virtual vehicle corresponding to each historical vehicle and the status information of each virtual vehicle ; Determine the fitting weight of each of the virtual vehicles based on historical traffic flow information, the motion information of the own vehicle, the state information of each of the virtual vehicles, and the grouping information of each of the virtual vehicles;
  • the fitting module 503 is configured to, based on the state information of each of the target vehicles, the fitting weight of each of the target vehicles, the state information of each of the virtual vehicles, and the fitting of each of the virtual vehicles.
  • the combined weights are fitted to generate one or more pieces of current traffic flow information.
  • the historical vehicle satisfies a cache condition; wherein, the cache condition is used to filter a vehicle traveling in the same direction as the vehicle and the vehicle does not belong to other groups.
  • the caching conditions include: the vehicle is located in front of the own vehicle and the relative distance from the vehicle is greater than a preset relative distance threshold; a difference between the direction of the speed of the vehicle relative to the own vehicle and the driving direction of the own vehicle The included angle is smaller than the preset included angle threshold; the lateral distance of the vehicle relative to the vehicle is within the preset lateral distance range.
  • the caching condition further includes: the life cycle of the vehicle is greater than a preset life cycle threshold.
  • the determination module 502 determines, for each of the target vehicles: a lateral speed deviation weight of the target vehicle, a lateral displacement of the target vehicle based on the state information of the target vehicle and the grouping information of the target vehicle The deviation weight, the life cycle weight of the target vehicle and the speed weight of the target vehicle; the lateral speed deviation weight of the target vehicle, the lateral displacement deviation weight of the target vehicle, the life cycle weight of the target vehicle and the target vehicle The speed weights are multiplied to obtain the fitting weights of the target vehicle.
  • the determining module 502 determines, for each of the virtual vehicles: a lateral speed deviation weight of the virtual vehicle, a lateral displacement of the virtual vehicle based on the state information of the virtual vehicle and the grouping information of the virtual vehicle Deviation weight, life cycle weight of the virtual vehicle, speed weight of the virtual vehicle and weight of the virtual vehicle; lateral speed deviation weight of the virtual vehicle, lateral displacement deviation weight of the virtual vehicle, life of the virtual vehicle The period weight, the speed weight of the virtual vehicle and the weight of the virtual vehicle are multiplied to obtain the fitting weight of the virtual vehicle.
  • each group corresponds to a traffic flow; for a group, the fitting module 503 selects a fitting method based on the number of target vehicles in the group and the longitudinal distribution distances of all target vehicles in the group; based on the selected The fitting method uses the motion information of the vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group to generate the current traffic flow information corresponding to the group by fitting.
  • the fitting module 503 selects the fitting method based on the number of target vehicles in the group and the longitudinal distribution distances of all target vehicles in the group, including: judging whether the number of target vehicles in the group satisfies the first quantity condition , and whether the longitudinal distribution distance is greater than the preset first distance threshold; if the first quantity condition is met and is greater than the preset first distance threshold, first-order fitting and second-order fitting are selected; Whether the number of target vehicles satisfies the second number condition, and whether the longitudinal distribution distance is greater than the preset second distance threshold; if it meets the second number condition and is greater than the preset second distance threshold, select first-order fitting; otherwise , no fitting is performed.
  • the fitting module 503 fitting and generating the current traffic flow information corresponding to the group includes: fitting and generating the first-order fitting result corresponding to the group, and determining the first mean square error of the first-order fitting result; generating the second-order fitting result corresponding to the group by fitting, and determining the second mean square error of the second-order fitting result; comparing the first mean square error and For the second mean square error, if the first mean square error is greater than a preset multiple of the second mean square error, select the second-order fitting result as the current traffic flow information corresponding to the group; otherwise, select the The first-order fitting result is the current traffic flow information corresponding to the group.
  • the fitting module 503 uses the motion information of the own vehicle, the state information of each target vehicle in the group, and the fitting weight of each target vehicle in the group to fit and generate the current corresponding to the group.
  • the traffic flow information includes: based on the state information of each target vehicle in the group and the fitting weight of each target vehicle in the group, fitting and generating the initial traffic flow information corresponding to the group;
  • the motion information of the vehicle is used to constrain the initial traffic flow information to obtain the current traffic flow information corresponding to the group.
  • the constraints include position constraints, heading constraints, and/or curvature constraints.
  • the location constraint includes: constraining the variation of the constant term in the fitting parameter corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information; the variation of the constant term The confidence levels corresponding to the historical traffic flow information are inversely correlated; the variation of the constant term is positively correlated with the fitting quantity corresponding to the initial traffic flow information.
  • the heading constraint includes: constraining a primary term in the fitting parameters corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information and/or the motion information of the host vehicle
  • the variation of the primary item is inversely correlated with the confidence level corresponding to the historical traffic flow information, and the variation of the primary item is positively correlated with the steering amplitude of the vehicle , and/or, there is a positive correlation between the variation of the primary term and the fitting quantity corresponding to the initial traffic flow information.
  • the curvature constraint includes: constraining the quadratic in the fitting parameters corresponding to the initial traffic flow information based on the confidence level corresponding to the historical traffic flow information and/or the motion information of the host vehicle The variation of the item; wherein, the variation of the quadratic item is inversely correlated with the confidence level corresponding to the historical traffic flow information, and the variation between the variation of the quadratic item and the speed of the vehicle is Inverse correlation, and/or, there is a positive correlation between the variation of the quadratic term and the rotational speed of the steering wheel of the host vehicle.
  • the apparatus for determining traffic flow information further includes a confidence level determination unit not shown in FIG. 5 , configured to: determine a confidence level increment based on the current traffic flow information; and based on historical traffic flow information The confidence level and the confidence level increment are determined to determine the confidence level corresponding to the current traffic flow information; wherein, the confidence level increment and the mean square error of the fitting result corresponding to the current traffic flow information are inversely correlated , there is a positive correlation between the confidence increment and the fitting quantity corresponding to the current traffic flow information, and/or, the confidence increment and the longitudinal distribution distance of the target vehicle corresponding to the current traffic flow information are positively correlated.
  • the confidence determination unit determines the confidence increment includes: if there is an intersection between the two current traffic flow information, comparing the fitting quantity and the longitudinal distribution distance corresponding to the two traffic flows; The confidence increment corresponding to the current traffic flow information with less or shorter longitudinal distribution distance is negative.
  • the apparatus for determining traffic flow information further includes a reference path determination unit not shown in FIG. 5 , configured to determine based on the current traffic flow information and a confidence level corresponding to the current traffic flow information
  • the driving reference path of the own vehicle; the determining the driving reference path of the own vehicle includes: if the confidence levels corresponding to the current traffic flow information on both sides of the own vehicle are all higher than the preset confidence threshold, then determining the parameters of the driving reference path of the own vehicle as this The weighted average of the fitting parameters corresponding to the current traffic flow information on both sides of the vehicle; wherein, the weight of the fitting parameters corresponding to the current traffic flow information on both sides of the vehicle is inversely correlated with the rate of change of the fitting parameters; if If the confidence corresponding to the current traffic flow information on only one side of the vehicle is higher than the preset confidence threshold, then determine the fitting parameters corresponding to the current traffic flow information on the side of the primary and quadratic terms in the parameters of the vehicle's driving reference path.
  • the primary term and the secondary term are the same, and the constant term in the parameters of the reference path of the vehicle is zero; if the confidence levels corresponding to the current traffic flow information on both sides of the vehicle are lower than the preset confidence threshold, it is determined that the vehicle is driving The parameters of the reference path are all zero.
  • the apparatus for determining traffic flow information further includes an auxiliary positioning marker determining unit not shown in FIG. 5 , configured to, based on the current traffic flow information and a confidence level corresponding to the current traffic flow information, Determining the auxiliary positioning mark; the determining the auxiliary positioning mark includes: if the confidence levels corresponding to the current traffic flow information on both sides of the vehicle are all higher than the preset confidence threshold, then determining the auxiliary positioning mark as the first mark; If the confidence corresponding to the current traffic flow information on the right side of the vehicle is higher than the preset confidence threshold, the auxiliary positioning mark is determined as the second mark; if the confidence corresponding to the current traffic flow information on the right side of the vehicle is higher than the preset confidence threshold, Then, it is determined that the auxiliary positioning flag is the third flag.
  • each unit in the apparatus for determining traffic flow information is only a logical function division, and in actual implementation, there may be other division methods, such as the grouping module 501 , the determining module 502 and the fitting module 503 . At least two units may be implemented as one unit; the grouping module 501 , the determining module 502 or the fitting module 503 may also be divided into multiple sub-units. It can be understood that each unit or sub-unit can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods for implementing the described functionality for each particular application.
  • the grouping information, fitting weight, motion information of the vehicle, and state information of the target vehicle may be used.
  • the embodiment of the present disclosure can determine the auxiliary positioning mark based on the current traffic flow information and the confidence level corresponding to the current traffic flow information.
  • the auxiliary positioning mark is independent of the lane line information. Assisting in determining the lane where the vehicle is located provides the necessary basis for the long-distance path planning of the L3-level automatic driving system.
  • the embodiments of the present disclosure can perform lateral control of the vehicle based on the current traffic flow information, get rid of the dependence on the lane line information, and help maintain the lateral control of the vehicle in a congested environment. Has industrial applicability.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种交通流信息的确定方法、装置、电子设备和存储介质，方法包括：基于本车的运动信息和一个或多个目标车辆的状态信息，对每个目标车辆进行分组，得到分组信息（401）；基于每个目标车辆的状态信息和分组信息，确定每个目标车辆的拟合权重（402）；基于本车的运动信息、每个目标车辆的状态信息、每个目标车辆的拟合权重和分组信息，拟合生成一条或多条当前交通流信息（403）。该方法通过对多个目标车辆进行分组，并确定每个目标车辆的拟合权重，进而可基于分组信息、拟合权重、本车的运动信息和目标车辆的状态信息，拟合生成当前交通流信息，实现对交通流的感知。

Description

一种交通流信息的确定方法、装置、电子设备和存储介质 技术领域
本公开实施例涉及智能驾驶技术领域，具体涉及一种交通流信息的确定方法、装置、电子设备和存储介质。
背景技术
当前针对结构化道路的智能驾驶系统，在进行行驶环境感知过程中，通常基于车辆前向摄像头采集的车道线信息进行车辆横向位置的定位与目标行驶路径的生成。可见，车辆横向定位(也即车辆所在车道的定位)、车辆横向控制和目标行驶路径生成的准确性依赖智能驾驶系统或摄像头对车道线的识别精度。
然而，目前存在多种因素导致车道线无法精确识别。例如，当路面交通较为拥堵，本车与前车车速较低、车距较近，两侧车道车流较密时，车辆前向摄像头的感知距离与视野范围会受到限制，此时根据摄像头采集的图像数据进行车道线识别将十分困难。另外，由于道路施工、车道变更、车流量过大等种种原因，经常会出现车道线不清晰、旧车道线清理不足等干扰车道线准确识别的情况。
可见，需要有其他感知手段降低车道线无法识别或识别准确度较低对车辆横向定位、车辆横向控制和目标行驶路径生成的影响。因此，亟需提供一种新的感知手段，能够感知交通流(Traffic Stream)信息，交通流可以理解为同一车道多个车辆同向运动所形成的车流，交通流信息可以理解为交通流的轨迹的信息。
发明内容
本公开提供一种新的感知手段，能够感知交通流信息。本公开的至少一个实施例提供了一种交通流信息的确定方法、装置、电子设备和非暂态计算机可读存储介质。
第一方面，本公开实施例提出的交通流信息的确定方法包括：
基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息；
基于每个所述目标车辆的状态信息和所述分组信息，确定每个所述目标车辆的拟合权重；
基于所述本车的运动信息、每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重和所述分组信息，拟合生成一条或多条当前交通流信息。
在一些实施例中，所述基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息包括：
基于所述本车的运动信息，确定本车的行驶轨迹；
基于所述本车的运动信息和每个所述目标车辆的状态信息，筛选一个或多个有效目标车辆；
基于所述本车的行驶轨迹和每个所述有效目标车辆的状态信息，对每个所述有效目标车辆进行分组，得到分组信息。
在一些实施例中，所述得到分组信息后，所述方法还包括：
基于历史交通流信息修正所述分组信息；
相应地，基于所述历史交通流信息、每个所述有效目标车辆的状态信息和修正后的分组信息，确定每个所述有效目标车辆的拟合权重。
在一些实施例中，所述基于所述本车的行驶轨迹和每个所述有效目标车辆的状态信息，对每个所述有效目标车辆进行分组，得到分组信息包括：
针对每个所述有效目标车辆：
基于该有效目标车辆的状态信息和所述本车的行驶轨迹，确定该有效目标车辆所在的相对车道；
基于所述相对车道，确定该有效目标车辆的分组信息。
在一些实施例中，所述分组信息包括以下五组：本车所在车道的左侧第一车道分组、本车所在车道的左侧第二车道分组、本车所在车道的右侧第一车道分组、本车所在车道的右侧第二车道分组和其他分组。
在一些实施例中，所述拟合生成一条或多条当前交通流信息之前，所述方法还包括：
判断目标车辆数量是否满足预设的拟合数量；若满足，则判断本车的行驶状态是否为直行；
若本车的行驶状态为直行，则获取缓存的历史车辆；
更新每个所述历史车辆的坐标，得到每个历史车辆对应的虚拟车辆及每个所述虚拟车辆的状态信息；
基于历史交通流信息、所述本车的运动信息、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的分组信息，确定每个所述虚拟车辆的拟合权重；
相应地，所述拟合生成一条或多条当前交通流信息包括：基于每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的拟合权重，拟合生成一条或多条当前交通流信息。
在一些实施例中，所述历史车辆满足缓存条件；其中，所述缓存条件用于筛选与本车同向行驶的车辆且该车辆不属于其他分组。
在一些实施例中，所述缓存条件包括：
车辆位于本车前方且与本车的相对距离大于预设的相对距离阈值；
车辆相对于本车的速度的方向与本车的行驶方向之间的夹角小于预设的夹角阈值；
车辆相对于本车的侧向距离处于预设的侧向距离范围内。
在一些实施例中，所述缓存条件还包括：车辆的生命周期大于预设的生命周期阈值。
在一些实施例中，所述确定每个所述目标车辆的拟合权重包括：
针对每个所述目标车辆：
基于该目标车辆的状态信息和该目标车辆的分组信息，确定该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重；
将该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重进行相乘，得到该目标车辆的拟合权重。
在一些实施例中，所述确定每个所述虚拟车辆的拟合权重包括：
针对每个所述虚拟车辆：
基于该虚拟车辆的状态信息和该虚拟车辆的分组信息，确定该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重；
将该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重进行相乘，得到该虚拟车辆的拟合权重。
在一些实施例中，每个分组对应一条交通流；所述拟合生成一条或多条当前交通流信息包括：
针对一个分组：
基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式；
基于选择的拟合方式，利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息。
在一些实施例中，所述基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式包括：
判断该分组中目标车辆的数量是否满足第一数量条件，且纵向分布距离是否大于预设的第一距离阈值；
若满足第一数量条件，且大于预设的第一距离阈值，则选择一阶拟合和二阶拟合；否则，判断该分组中目标车辆的数量是否满足第二数量条件，且纵向分布距离是否大于预设的第二距离阈值；
若满足第二数量条件，且大于预设的第二距离阈值，则选择一阶拟合；否则，不进行拟合。
在一些实施例中，若选择一阶拟合和二阶拟合，则所述拟合生成该分组对应的当前交通流信息包括：
拟合生成该分组对应的一阶拟合结果，并确定所述一阶拟合结果的第一均方差；
拟合生成该分组对应的二阶拟合结果，并确定所述二阶拟合结果的第二均方差；
比较所述第一均方差和所述第二均方差，若所述第一均方差大于所述第二均方差的预设倍数，则选择所述二阶拟合结果为该分组对应的当前交通流信息；否则，选择所述一阶拟合结果为该分组对应的当前交通流信息。
在一些实施例中，所述利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息包括：
基于该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的初始交通流信息；
基于历史交通流信息和所述本车的运动信息，对所述初始交通流信息进行约束，得到该分组对应的当前交通流信息。
在一些实施例中，所述约束包括：位置约束、航向约束和/或曲率约束。
在一些实施例中，所述位置约束包括：基于所述历史交通流信息对应的置信度约束所述初始交通流信息对应的拟合参数中的常数项的变化量；
所述常数项的变化量与所述历史交通流信息对应的置信度之间为反相关；所述常数项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
在一些实施例中，所述航向约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的一次项的变化量；
其中，所述一次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述一次项的变化量与所述本车的转向幅度之间为正相关，和/或，所述一次项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
在一些实施例中，所述曲率约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的二次项的变化量；
其中，所述二次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述二次项的变化量与所述本车的速度之间为反相关，和/或，所述二次项的变化量与所述本车的方向盘转速之间为正相关。
在一些实施例中，所述方法还包括：
基于所述当前交通流信息，确定置信度增量；
基于历史交通流信息的置信度和所述置信度增量，确定所述当前交通流信息对应的置信度；
其中，所述置信度增量与所述当前交通流信息对应的拟合结果均方差之间为反相关，所述置信度增量与所述当前交通流信息对应的拟合数量之间为正相关，和/或，所述置信度增量与所述当前交通流信息对应的目标车辆的纵向分布距离之间为正相关。
在一些实施例中，所述基于所述当前交通流信息，确定置信度增量包括：
若两条当前交通流信息存在交叉，则比较两条交通流对应的拟合数量和纵向分布距离；确定拟合数量较少或纵向分布距离较短的当前交通流信息对应的置信度增量为负。
在一些实施例中，所述方法还包括：基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定本车行驶参考路径；
所述确定本车行驶参考路径包括：
若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定本车行驶参考路径的参数为本车两侧的当前交通流信息对应的拟合参数的加权平均值；其中，本车两侧的当前交通流信息对应的拟合参数的权重与拟合参数的变化率之间为反相关；
若本车仅一侧的当前交通流信息对应的置信度高于预设置信度门限，则确定本车行驶参考路径的参数中一次项和二次项与该侧当前交通流信息对应的拟合参数中一次项和二次项相同，本车行驶参考路径的参数中常数项为零；
若本车两侧的当前交通流信息对应的置信度均低于预设置信度门限，则确定本车行驶参考路径的参数均为零。
在一些实施例中，所述方法还包括：基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定辅助定位标志；
所述确定辅助定位标志包括：
若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定辅助定位标志为第一标志；
若本车左侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第二标志；
若本车右侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第三标志。
第二方面，本公开实施例提供的交通流信息的确定装置包括：
分组模块，用于基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息；
确定模块，用于基于每个所述目标车辆的状态信息和所述分组信息，确定每个所述目标车辆的拟合权重；
拟合模块，用于基于所述本车的运动信息、每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重和所述分组信息，拟合生成一条或多条当前交通流信息。
第三方面，本公开实施例提供的电子设备包括：处理器和存储器；所述处理器通过调用所述存储器存储的程序或指令，用于执行如第一方面任一实施例所述交通流信息的确定方法的步骤。
第四方面，本公开实施例提供的非暂态计算机可读存储介质，用于存储程序或指令，所述程序或指令使计算机执行如第一方面任一实施例所述交通流信息的确定方法的步骤。
可见，本公开的至少一个实施例中，通过对多个目标车辆进行分组，并确定每个目标车辆的拟合权重，进而可基于分组信息、拟合权重、本车的运动信息和目标车辆的状态信息，拟合生成当前交通流信息，实现对交通流的感知。
附图说明
为了更清楚地说明本公开实施例的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本公开的一些实施例，对于本领域普通技术人员来讲，还可以根据这些附图获得其他的附图。
图1是本公开实施例提供的一种确定交通流信息的示例性场景图；
图2是本公开实施例提供的一种智能驾驶车辆的示例性架构图；
图3是本公开实施例提供的一种电子设备的示例性框图；
图4是本公开实施例提供的一种交通流信息的确定方法的示例性流程图；
图5是本公开实施例提供的一种交通流信息的确定装置的示例性框图。
具体实施方式
为了能够更清楚地理解本公开的上述目的、特征和优点，下面结合附图和实施例对本公开作进一步的详细说明。可以理解的是，所描述的实施例是本公开的一部分实施例，而不是全部的实施例。此处所描述的具体实施例仅仅用于解释本公开，而非对本公开的限定。基于所描述的本公开的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本公开保护的范围。
需要说明的是，在本文中，诸如“第一”和“第二”等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。
随着智能驾驶技术的发展，远距离的路径规划将成为L3等更高级别自动驾驶系统的必备能力，这种规划能力需要对本车所在车道进行精确定位。以目前的低成本卫星定位系统还无法准确定位本车所在车道，智能驾驶系统或摄像头对车道线类型的识别亦不能保证对当前车道的准确判断，况且目前还存在多种因素导致车道线无法精确识别，例如背景技术中提及的因素，在此不再赘述。
因此，本公开实施例提供一种新的感知手段，能够感知交通流信息，交通流可以理解为同一车道多个车辆同向运动所形成的车流，交通流信息可以理解为交通流的轨迹的信息。由于车道线可以看作是一条线，而轨迹也可以看作是一条线，因此，交通流信息可以替代车道线，用于车辆横向定位(也即车辆所在车道的定位)、车辆横向控制和目标行驶路径生成。
本公开实施例提供了一种交通流信息的确定方法、装置、电子设备或存储介质，通过对多个目标车辆进行分组，并确定每个目标车辆的拟合权重，进而可基于分组信息、拟合权重、本车的运动信息和目标车辆的状态信息，拟合生成当前交通流信息，实现对交通流的感知。本公开实施例可以应用于智能驾驶车辆，还可以应用于电子设备。所述智能驾驶车辆为搭载不同等级智能驾驶系统的车辆，智能驾驶系统例如包括：无人驾驶系统、辅助驾驶系统、驾驶辅助系统、高度自动驾驶系统、完全自动驾驶车辆等等。所述电子设备安装有智能驾驶系统，例如电子设备可用于测试智能驾驶算法，又例如电子设备可以为车载设备，在一些实施例中，电子设备还可以应用到其他领域。以下为了能够更清楚无误的阐述，本公开实施例以智能驾驶车辆为例对所述交通流信息的确定方法、装置、电子设备或存储介质进行说明。
图1是本公开实施例提供的一种确定交通流信息的示例性场景图，图1中，本车101正在直行，本车轨迹的左右边界为104所示，在本车左侧有3辆车，例如图1中103所示，右侧有2辆车，期望通过本公开实施例提供的交通流信息的确定方法，确定出本车左侧交通流信息，例如图1中左侧交通流对应的轨迹102，以及确定出本车右侧交通流信息。
图2为本公开实施例提供的一种智能驾驶车辆的示例性整体架构图。图2所示的智能驾驶车辆可以实现为图1中的车辆101。如图2所示，智能驾驶车辆包括：传感器组、智能驾驶系统、车辆底层执行系统以及其他可用于驱动车辆和控制车辆运行的部件，例如制动踏板、方向盘和油门踏板。
传感器组，用于采集车辆外界环境的数据和探测车辆的位置数据。传感器组例如包括但不限于摄像头、激光雷达、毫米波雷达、超声波雷达、GPS(Global Positioning System，全球定位系统)和IMU(Inertial Measurement Unit，惯性测量单元)中的至少一个。
在一些实施例中，传感器组，还用于采集车辆的动力学数据，传感器组例如还包括但不限于车轮转速传感器、速度传感器、加速度传感器、方向盘转角传感器、前轮转角传感器中的至少一个。
智能驾驶系统，用于获取传感器组的传感数据，其中，所述传感数据包括但不限于图像、视频、激光点云、毫米波、GPS信息、车辆状态等。在一些实施例中，智能驾驶系统基于所述传感数据进行环境感知和车辆定位，生成感知信息和车辆位姿；基于所述感知信息和车辆位姿进行规划和决策，生成规划和决策信息；基于规划和决策信息生成车辆控制指令，并下发给车辆底层执行系统。其中，控制指令可包括但不限于：方向盘转向、横向控制指令、纵向控制指令等。
在一些实施例中，智能驾驶系统获取传感器数据、V2X(Vehicle to X，车用无线通信)数据、高 精度地图等数据并基于以上至少一种数据进行环境感知与定位，生成感知信息和定位信息。其中，感知信息可包括但不限于以下至少一个：障碍物信息、道路标志/标记、行人/车辆信息、可行驶区域。定位信息包括车辆位姿。
在一些实施例中，智能驾驶系统基于感知信息和车辆位姿，以及V2X数据、高精度地图等数据中的至少一种，生成规划和决策信息。其中，规划信息可包括但不限于规划路径等；决策信息可包括但不限于以下至少一种：行为(例如包括但不限于跟车、超车、停车、绕行等)、车辆航向、车辆速度、车辆的期望加速度、期望的方向盘转角等。
在一些实施例中，本公开实施例所提供的交通流信息的确定方法可应用于所述智能驾驶系统中。
在一些实施例中，智能驾驶系统可以为软件系统、硬件系统或者软硬件结合的系统。例如，智能驾驶系统是运行在操作系统上的软件系统，车载硬件系统是支持操作系统运行的硬件系统。
在一些实施例中，智能驾驶系统可以与云端服务器进行交互。在一些实施例中，智能驾驶系统与云端服务器通过无线通讯网络(例如包括但不限于GPRS网络、Zigbee网络、Wifi网络、3G网络、4G网络、5G网络等无线通讯网络)进行交互。
在一些实施例中，云端服务器用于与车辆进行交互。其中，所述云端服务器可以向车辆发送环境信息、定位信息、控制信息及车辆智能驾驶过程中需要的其他信息。在一些实施例中，所述云端服务器可以接收来自车端的传感数据、车辆状态信息、车辆行驶信息以及车辆请求的相关信息。在一些实施例中，云端服务器可以基于用户设置或车辆请求对所述车辆进行远程控制。在一些实施例中，云端服务器可以是一个服务器，也可以是一个服务器群组。服务器群组可以是集中式的，也可以是分布式的。在一些实施例中，云端服务器可以是本地的或远程的。
车辆底层执行系统，用于接收车辆控制指令，并基于所述车辆控制指令控制车辆行驶。在一些实施例中，车辆底层执行系统包括但不限于：转向系统、制动系统和驱动系统。在一些实施例中，所述车辆底层执行系统还可包括底层控制器，用于可以解析车辆控制指令，并将其分别下发至转向系统、制动系统和驱动系统等对应系统。
在一些实施例中，智能驾驶车辆还可包括图2中未示出的车辆CAN总线，车辆CAN总线连接车辆底层执行系统。智能驾驶系统与车辆底层执行系统之间的信息交互通过车辆CAN总线进行传递。
图3是本公开实施例提供的一种电子设备的结构示意图。在一些实施例中，电子设备可以为车载设备。在一些实施例中，电子设备可支持智能驾驶系统的运行。
如图3所示，电子设备包括：至少一个处理器301、至少一个存储器302和至少一个通信接口303。车载设备中的各个组件通过总线系统304耦合在一起。通信接口303，用于与外部设备之间的信息传输。可理解地，总线系统304用于实现这些组件之间的连接通信。总线系统304除包括数据总线之外，还包括电源总线、控制总线和状态信号总线。但为了清楚说明起见，在图3中将各种总线都标为总线系统304。
可以理解，本实施例中的存储器302可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。
在一些实施方式中，存储器302存储了如下的元素，可执行单元或者数据结构，或者他们的子集， 或者他们的扩展集：操作系统和应用程序。
其中，操作系统，包含各种系统程序，例如框架层、核心库层、驱动层等，用于实现各种基础任务以及处理基于硬件的任务。应用程序，包含各种应用程序，例如媒体播放器(Media Player)、浏览器(Browser)等，用于实现各种应用任务。实现本公开实施例提供的交通流信息的确定方法的程序可以包含在应用程序中。
在本公开实施例中，处理器301通过调用存储器302存储的程序或指令，具体的，可以是应用程序中存储的程序或指令，处理器301用于执行本公开实施例提供的交通流信息的确定方法各实施例的步骤。
本公开实施例提供的交通流信息的确定方法可以应用于处理器301中，或者由处理器301实现。处理器301可以是一种集成电路芯片，具有信号的处理能力。在实现过程中，上述方法的各步骤可以通过处理器301中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器301可以是通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
本公开实施例提供的交通流信息的确定方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件单元组合执行完成。软件单元可以位于随机存储器，闪存、只读存储器，可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器302，处理器301读取存储器302中的信息，结合其硬件完成方法的步骤。
图4为本公开实施例提供的一种交通流信息的确定方法的示例性流程图。该方法的执行主体为电子设备，在一些实施例中，该方法的执行主体还可以为电子设备所支持的智能驾驶系统。为便于描述，以下实施例中以智能驾驶系统为执行主体说明交通流信息的确定方法的流程。
如图4所示，在步骤401中，智能驾驶系统基于本车的运动信息和一个或多个目标车辆的状态信息，对每个目标车辆进行分组，得到分组信息。
在一些实施例中，本车的运动信息为本车在行驶过程中的与行驶相关的动力学信息。例如，本车的运动信息可以包括但不限于车轮转速、车速、加速度、方向盘转角、前轮转角、横摆角速度。在一些实施例中，本车的运动信息可以通过图1所示的传感器组来采集，智能驾驶系统可从传感器组获取本车的运动信息。
在一些实施例中，智能驾驶系统获取传感器数据、V2X数据、高精度地图等数据并基于以上至少一种数据进行环境感知，可得到一个或多个目标车辆的状态信息。
在一些实施例中，目标车辆的状态信息为目标车辆与本车之间的相对位置和相对运动等信息的总称。例如，目标车辆的状态信息可以包括但不限于目标车辆相对于本车的侧向位移、纵向位移、侧向速度和纵向速度等中的至少一个，其中，侧向可以理解为横向，也即垂直于车道线方向。在一些实施例中，目标车辆的状态信息还可以包括但不限于标识(ID)、感知类型、生命周期和目标置信度等中的一个或多个。其中，生命周期可以理解为目标车辆首次被感知的时刻至当前时刻之间的时长，也即， 目标车辆出现在本车“视野”内的时长。目标置信度用于表示目标车辆的可信程度，由智能驾驶系统的感知算法自动生成。
在一些实施例中，在对每个目标车辆进行分组时，智能驾驶系统所采用的分组规则为将目标车辆的分组按照其所在车道分为五组：本车所在车道的左侧第一车道分组、本车所在车道的左侧第二车道分组、本车所在车道的右侧第一车道分组、本车所在车道的右侧第二车道分组和其他分组。其中，其他分组为前述四个分组以外的任意情形。例如，其他分组可以为本车所在车道、本车所在车道的左侧第三车道和本车所在车道的右侧第三车道中的一个或多个。
在一些实施例中，在对每个目标车辆进行分组后，智能驾驶系统可对每个目标车辆进行分组标记，分组标记用于表征目标车辆所属的分组，例如，分组标记可以为分组标志位，该分组标志位的不同取值表示不同分组。本实施例中，分组信息至少包括分组标记。在一些实施例中，分组信息还可以包括但不限于目标车辆的ID、感知类型、生命周期和目标置信度等中的一个或多个。
在一些实施例中，在对每个目标车辆进行分组时，智能驾驶系统基于本车的运动信息，确定本车的行驶轨迹；进而基于本车的行驶轨迹和每个目标车辆的状态信息，可确定每个目标车辆所在的相对车道(相对车道是相对于本车所在车道而言)；从而可按照前述分组规则，基于每个目标车辆所在的相对车道，对每个目标车辆进行分组，得到分组信息。
在一些实施例中，在确定本车的行驶轨迹时，智能驾驶系统基于本车的车速、方向盘转角和横摆角速度的信号经过滤波后计算本车的转弯半径，进而基于转弯半径确定本车当前的行驶轨迹。需要说明的是，由转弯半径确定行驶轨迹属于智能驾驶领域的成熟技术，在此不再赘述具体的确定过程。
在一些实施例中，考虑到目标车辆可能为无效车辆，例如目标车辆与本车反向而行，不应参与后续拟合交通流信息，因此，为了避免无效车辆参与后续拟合交通流信息，导致拟合准确性降低，本实施例中，在进行分组之前，智能驾驶系统首先筛选有效目标车辆，过滤无效目标车辆；然后基于本车的行驶轨迹和每个有效目标车辆的状态信息，对每个有效目标车辆进行分组，得到分组信息。
在一些实施例中，在筛选有效目标车辆时，智能驾驶系统基于本车的运动信息和每个目标车辆的状态信息，筛选一个或多个有效目标车辆。例如，智能驾驶系统采用去除无效目标车辆的方式进行筛选，无效目标车辆包括：纵向速度为负(本车的速度为正)的目标车辆或纵向距离大于预设的纵向距离阈值的目标车辆。也即，有效目标车辆的纵向速度为正(或为零)且纵向距离小于或等于预设的纵向距离阈值。
在一些实施例中，考虑到对目标车辆分组时，本车可能正在变道、转向等非直行运动，可能造成分组错误，例如，目标车辆在本车左侧第一车道保持直行，应当将目标车辆划分到本车所在车道的左侧第一车道分组，然而，本车向右快速转向时，可能将该目标车辆划分到本车所在车道的左侧第二车道分组，造成分组错误。因此，需要对分组信息进行修正，避免本车快速转向造成分组错误。
在一些实施例中，在对分组信息进行修正时，智能驾驶系统基于历史交通流信息修正分组信息。历史交通流信息是上一个确定周期拟合生成的，其中，确定周期是智能驾驶系统预先设置的确定交通流信息的周期，在实际应用中，可基于应用场景对拟合结果刷新率的要求，设置确定周期，在几十毫秒至几秒之间取值，例如设置确定周期为5秒。每个分组对应一条交通流，也即，一条历史交通流信 息与一个分组对应。其中，历史交通流信息包括拟合参数的值，该拟合参数包括二次项、一次项和常数项。若二次项存在，则历史交通流信息对应一条曲线，若二次项为零，一次项不为零，则历史交通流信息对应一条直线。在一些实施例中，历史交通流信息还包括置信度，该置信度用于表示交通流的可信程度。若没有历史交通流信息，例如，本车由车库驶出并进入低速车道时，智能驾驶系统中还没有存储历史交通流信息，那么，历史交通流信息的置信度为零，不修正分组信息。
在一些实施例中，在基于历史交通流信息修正分组信息时，智能驾驶系统判断该历史交通流信息周围预设范围内是否存在目标车辆，若存在，则将该目标车辆的分组修正为该历史交通流信息对应的分组。其中，预设范围可根据实际需要进行设置，本实施例不限定预设范围的具体取值。例如，某个目标车辆的分组为左侧第二车道分组，且该目标车辆位于历史交通流信息周围预设范围内，而该历史交通流信息对应左侧第一车道分组，因此，将该目标车辆的分组由左侧第二车道分组修正为左侧第一车道分组。
在一些实施例中，在基于历史交通流信息修正分组信息后，智能驾驶系统将修正后的各分组中满足缓存条件的车辆进行缓存，对于下一个确定周期，缓存的车辆即为历史车辆。在一些实施例中，可以设置缓存时长，缓存时长可基于实际需要进行设置，本实施例不限定缓存时长的具体取值。
其中，缓存条件用于筛选与本车同向行驶的车辆且该车辆不属于其他分组。需要说明的是，针对每个确定周期，智能驾驶系统会将该确定周期中修正后的各分组中满足缓存条件的车辆进行缓存，也即缓存的时机是在该确定周期内。
在步骤402中，智能驾驶系统基于每个目标车辆的状态信息和分组信息，确定每个目标车辆的拟合权重。
在一些实施例中，在确定每个目标车辆的拟合权重时，智能驾驶系统针对每个目标车辆：基于该目标车辆的状态信息和该目标车辆的分组信息，确定该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重；进而将该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重进行相乘，得到该目标车辆的拟合权重。
其中，该目标车辆的侧向速度偏离权重的计算方式为：计算该目标车辆所属分组中所有目标车辆的平均侧向速度；基于该目标车辆的侧向速度和所述平均侧向速度，计算该目标车辆的侧向速度偏离权重。本实施例中，若该目标车辆的侧向速度偏离平均侧向速度越大，则侧向速度偏离权重越低。
该目标车辆的侧向位移偏离权重的计算方式为：计算该目标车辆所属分组中所有目标车辆的平均侧向位移；基于该目标车辆的侧向位移和所述平均侧向位移，计算该目标车辆的侧向位移偏离权重。本实施例中，若该目标车辆的侧向位移偏离平均侧向位移越远，则侧向位移偏离权重越低。
该目标车辆的生命周期权重的计算方式为：基于该目标车辆的生命周期和/或目标置信度，计算该目标车辆的生命周期权重。本实施例中，该目标车辆的生命周期越长，生命周期权重越高。该目标车辆的目标置信度越高，生命周期权重越高。
该目标车辆的速度权重的计算方式为：若不存在历史交通流信息与该目标车辆的分组相对应，则该目标车辆的速度权重为1。
在一些实施例中，可结合历史交通流信息来确定目标车辆的侧向位移偏离权重。具体地，基于该目标车辆的侧向位移和该目标车辆的分组所对应的历史交通流信息，计算该目标车辆的侧向位移偏离权重。本实施例中，若该目标车辆的侧向位移偏离历史交通流越远，则侧向位移偏离权重越低。
在一些实施例中，可结合历史交通流信息来确定目标车辆的速度权重。具体地，基于该目标车辆的侧向速度和纵向速度，合成该目标车辆的行驶速度，进而确定该目标车辆的行驶方向；从而基于该目标车辆的行驶方向和该目标车辆的分组所对应的历史交通流信息，计算该目标车辆的速度权重。本实施例中，若该目标车辆的行驶方向与历史交通流的切线方向之间的夹角越大，该目标车辆的速度权重越低。
在一些实施例中，在对每个有效目标车辆进行分组，得到分组信息，且利用历史交通流信息修正分组信息后，智能驾驶系统基于历史交通流信息、每个有效目标车辆的状态信息和修正后的分组信息，确定每个有效目标车辆的拟合权重。
在步骤403中，智能驾驶系统基于本车的运动信息、每个目标车辆的状态信息、每个目标车辆的拟合权重和分组信息，拟合生成一个或多个当前交通流信息。
在一些实施例中，每个分组对应一条交通流，针对一个分组，智能驾驶系统拟合生成该分组对应的当前交通流信息，具体地，基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式；进而基于选择的拟合方式，利用本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息。左侧第一车道分组拟合得到的交通流称为本车左侧交通流；右侧第一车道分组拟合得到的交通流称为本车右侧交通流。
在一些实施例中，在选择拟合方式时，智能驾驶系统判断该分组中目标车辆的数量是否满足第一数量条件，且纵向分布距离是否大于预设的第一距离阈值；其中，第一数量条件例如为3，第一距离阈值例如为50米，避免过拟合(例如，交通流实际为直线，但是采用曲线方式进行拟合)，第一数量条件和第一距离阈值的具体取值可根据实际需要进行设置。
若满足第一数量条件，且大于预设的第一距离阈值，则选择一阶拟合和二阶拟合；否则，判断该分组中目标车辆的数量是否满足第二数量条件，且纵向分布距离是否大于预设的第二距离阈值；其中，其中，第二数量条件例如为大于等于2且小于4，第二距离阈值例如为30米，第一数量条件和第一距离阈值的具体取值可根据实际需要进行设置。
若满足第二数量条件，且大于预设的第二距离阈值，则选择一阶拟合；否则，不进行拟合，延续上一交通流信息确定周期的数据。
在一些实施例中，若选择一阶拟合和二阶拟合，则智能驾驶系统拟合生成该分组对应的当前交通流信息时，拟合生成该分组对应的一阶拟合结果，并确定一阶拟合结果的第一均方差；拟合生成该分组对应的二阶拟合结果，并确定二阶拟合结果的第二均方差；进而比较第一均方差和第二均方差，若第一均方差大于第二均方差的预设倍数(例如1.5倍，还可以根据实际情况设置预设倍数)，则选择二阶拟合结果为该分组对应的当前交通流信息；否则，选择一阶拟合结果为该分组对应的当前交通流信息。
在一些实施例中，利用本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息时，智能驾驶系统基于该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的初始交通流信息；进而基于历史交通流信息和本车的运动信息，对初始交通流信息进行约束，得到该分组对应的当前交通流信息。
在一些实施例中，对初始交通流信息进行约束包括：位置约束、航向约束和/或曲率约束。
在一些实施例中，位置约束包括：基于历史交通流信息对应的置信度约束初始交通流信息对应的拟合参数中的常数项的变化量；其中，常数项的变化量与历史交通流信息对应的置信度之间为反相关；常数项的变化量与初始交通流信息对应的拟合数量之间为正相关。其中，拟合数量可以理解为参与拟合的目标车辆的数量。
在一些实施例中，航向约束包括：基于历史交通流信息对应的置信度和/或本车的运动信息，约束初始交通流信息对应的拟合参数中的一次项的变化量；其中，一次项的变化量与历史交通流信息对应的置信度之间为反相关，一次项的变化量与本车的转向幅度之间为正相关，和/或，一次项的变化量与初始交通流信息对应的拟合数量之间为正相关。
在一些实施例中，曲率约束包括：基于历史交通流信息对应的置信度和/或本车的运动信息，约束初始交通流信息对应的拟合参数中的二次项的变化量；其中，二次项的变化量与历史交通流信息对应的置信度之间为反相关，二次项的变化量与本车的速度之间为反相关，和/或，二次项的变化量与本车的方向盘转速之间为正相关。
在一些实施例中，每个分组对应一条交通流，针对一个分组，在拟合生成该分组对应的当前交通流信息之前，例如在对每个目标车辆进行分组，得到分组信息后，智能驾驶系统要确保该分组中目标车辆数量满足预设的拟合数量，否则无法进行拟合。因此，智能驾驶系统判断目标车辆数量是否满足预设的拟合数量；若满足，则判断本车的行驶状态是否为直行；若为直行，则获取缓存的历史车辆，且获取的历史车辆可以是一个或多个，且获取的历史车辆属于该分组。
在一些实施例中，历史车辆满足缓存条件；所述缓存条件用于筛选与本车同向行驶的车辆且该车辆不属于其他分组。
在一些实施例中，缓存条件包括以下(1)至(3)：
(1)车辆位于本车前方且与本车的相对距离大于预设的相对距离阈值；其中，预设的相对距离阈值例如为20米，本实施例不限定相对距离阈值的具体取值，本领域技术人员可根据实际需要进行设置。可以理解，判断一个车辆是否满足缓存条件时，是基于判断时刻该车辆与本车之间的相对关系，包括相对距离、行驶方向、侧向距离等。
(2)车辆相对于本车的速度的方向与本车的行驶方向之间的夹角小于预设的夹角阈值。其中，该车辆相对于本车的速度的方向可以理解为该车辆的行驶方向。该车辆的行驶方向可基于该车辆在判断时刻的侧向速度和纵向速度，合成该车辆的速度(也即行驶速度)，进而确定该车辆的行驶方向。
(3)车辆相对于本车的侧向距离处于预设的侧向距离范围内。其中，预设的侧向距离范围例如为2米至3米，本实施例不限定侧向距离范围的具体取值，本领域技术人员可根据实际需要进行设置。
在一些实施例中，缓存条件还可包括以下(4)：
(4)车辆的生命周期大于预设的生命周期阈值。其中，预设的生命周期阈值例如为3秒，本实施例不限定生命周期阈值的具体取值，本领域技术人员可根据实际需要进行设置。
在一些实施例中，在获取缓存的历史车辆后，智能驾驶系统更新每个历史车辆的坐标，得到每个历史车辆对应的虚拟车辆及每个虚拟车辆的状态信息。更新坐标的方式例如为：计算该历史车辆的缓存时刻与当前时刻的时差，利用历史车辆的状态信息和计算的时差，更新坐标，具体地，利用历史车辆的状态信息中的侧向速度和纵向速度，合成历史车辆的行驶速度，基于该行驶速度和计算的时差，更新坐标。其中，历史车辆的状态信息是该历史车辆在其缓存时所对应的状态信息。
在一些实施例中，虚拟车辆的状态信息包括但不限于虚拟车辆相对于本车的侧向位移、纵向位移、侧向速度和纵向速度等中的至少一个。其中，侧向位移、纵向位移基于前述坐标更新来确定，侧向速度和纵向速度与对应的历史车辆的侧向速度和纵向速度相同。在一些实施例中，虚拟车辆的状态信息还可以包括但不限于标识(ID)、感知类型、生命周期和目标置信度等中的一个或多个，这些信息与该虚拟车辆对应的历史车辆相同，例如，虚拟车辆的感知类型为对应的历史车辆的感知类型。虚拟车辆的分组信息与对应的历史车辆的分组信息相同。
在一些实施例中，若本车的行驶状态不为直行，则不筛选历史车辆，因为无法确定历史车辆与本车的相对关系，也无法将历史车辆的坐标更新得到虚拟车辆，即使将历史车辆的坐标更新得到虚拟车辆，但由于历史车辆与本车的相对关系不确定，因此，虚拟车辆与本车的相对关系也不确定，因而无法确定虚拟车辆的状态信息。
在一些实施例中，在确定虚拟车辆及虚拟车辆的状态信息后，智能驾驶系统可基于历史交通流信息、本车的运动信息、每个虚拟车辆的状态信息和每个虚拟车辆的分组信息，确定每个虚拟车辆的拟合权重。
在一些实施例中，在确定每个虚拟车辆的拟合权重时，智能驾驶系统针对每个虚拟车辆：基于该虚拟车辆的状态信息和该虚拟车辆的分组信息，确定该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重；进而将该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重进行相乘，得到该虚拟车辆的拟合权重。
其中，该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重和该虚拟车辆的速度权重的计算方式分别与前述目标车辆的侧向速度偏离权重、目标车辆的侧向位移偏离权重、目标车辆的生命周期权重和目标车辆的速度权重的计算方式相同，不再赘述。而虚拟车辆的权重基于该虚拟车辆的缓存时长进行计算。该虚拟车辆的缓存时长越长，虚拟车辆的权重越低。
在一些实施例中，在确定虚拟车辆及虚拟车辆的状态信息后，智能驾驶系统基于每个目标车辆的状态信息、每个目标车辆的拟合权重、每个虚拟车辆的状态信息和每个虚拟车辆的拟合权重，拟合生成一条或多条当前交通流信息。拟合生成的过程类似于前述仅基于目标车辆进行拟合的过程，在此不再赘述。
在一些实施例中，在拟合生成当前交通流信息后，智能驾驶系统可基于当前交通流信息，确定置信度增量；进而基于历史交通流信息的置信度和置信度增量，确定当前交通流信息对应的置信度。例如，将基于历史交通流信息的置信度与置信度增量相加，得到当前交通流信息对应的置信度。需要说明的是，当前交通流信息与历史交通流信息均对应同一分组，才能采用本实施例的方式确定当前交通流信息对应的置信度。
在一些实施例中，置信度增量与当前交通流信息对应的拟合结果均方差之间为反相关，置信度增量与当前交通流信息对应的拟合数量之间为正相关，和/或，置信度增量与当前交通流信息对应的目标车辆(若存在虚拟车辆，则虚拟车辆也应考虑在内)的纵向分布距离之间为正相关。
在一些实施例中，在确定置信度增量时，智能驾驶系统判断是否存在两条当前交通流信息交叉，若存在，则比较两条交通流对应的拟合数量和纵向分布距离；确定拟合数量较少或纵向分布距离较短的当前交通流信息对应的置信度增量为负。例如，智能驾驶系统判断本车左侧交通流和本车右侧交通流是否存在交叉，若存在，则比较两条交通流对应的拟合数量和纵向分布距离；确定拟合数量较少或纵向分布距离较短的当前交通流信息对应的置信度增量为负。
在一些实施例中，若本次交通流信息确定周期未进行拟合，则确定置信度增量为负。这个负值会累计到对应的历史交通流信息的置信度。
在一些实施例中，智能驾驶系统可基于当前交通流信息和当前交通流信息对应的置信度，确定本车行驶参考路径。确定本车行驶参考路径包括：
若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定本车行驶参考路径的参数为本车两侧的当前交通流信息对应的拟合参数的加权平均值；其中，本车两侧的当前交通流信息对应的拟合参数的权重与拟合参数的变化率之间为反相关；
若本车仅一侧的当前交通流信息对应的置信度高于预设置信度门限，则确定本车行驶参考路径的参数中一次项和二次项与该侧当前交通流信息对应的拟合参数中一次项和二次项相同，本车行驶参考路径的参数中常数项为零；
若本车两侧的当前交通流信息对应的置信度均低于预设置信度门限，则确定本车行驶参考路径的参数均为零。
在一些实施例中，智能驾驶系统可基于当前交通流信息和当前交通流信息对应的置信度，确定辅助定位标志。其中，辅助定位标志用于辅助确定本车所在车道。辅助定位标志独立于车道线信息，可在部分多车道场景，车道线不足以确定唯一车道时，辅助确定本车所在车道，为L3级自动驾驶系统的远距离路径规划提供了必要依据。
在一些实施例中，确定辅助定位标志包括：
若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定辅助定位标志为第一标志。第一标志代表本车两侧均存在车道。辅助定位标志可以为标志位，第一标志对应的标志位取值例如为1，当智能驾驶系统确定辅助定位标志为1时，确定本车两侧均存在车道。
若本车左侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第二标志。第二标志代表本车左侧存在车道。第二标志对应的标志位取值例如为2。
若本车右侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第三标志。第三标志代表本车右侧存在车道。第三标志对应的标志位取值例如为3。
在一些实施例中，智能驾驶系统可基于当前交通流信息进行本车横向控制，摆脱了对车道线信息的依赖，有利于在拥堵环境下维持本车的横向控制。基于一条路径进行横向控制属于本领域成熟技术手段，在此不再赘述。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员能够理解，本公开实施例并不受所描述的动作顺序的限制，因为依据本公开实施例，某些步骤可以采用其他顺序或者同时进行。另外，本领域技术人员能够理解，说明书中所描述的实施例均属于可选实施例。
本公开实施例还提出一种非暂态计算机可读存储介质，所述非暂态计算机可读存储介质存储程序或指令，所述程序或指令使计算机执行如交通流信息的确定方法各实施例的步骤，为避免重复描述，在此不再赘述。
图5是本公开实施例提供的一种交通流信息的确定装置的示例性框图。如图5所示，交通流信息的确定装置包括但不限于分组模块501、确定模块502和拟合模块503。
在图5中，分组模块501，用于基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息。
确定模块502，用于基于每个所述目标车辆的状态信息和所述分组信息，确定每个所述目标车辆的拟合权重。
拟合模块503，用于基于所述本车的运动信息、每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重和所述分组信息，拟合生成一条或多条当前交通流信息。
在一些实施例中，分组模块501，用于基于所述本车的运动信息，确定本车的行驶轨迹；基于所述本车的运动信息和每个所述目标车辆的状态信息，筛选一个或多个有效目标车辆；基于所述本车的行驶轨迹和每个所述有效目标车辆的状态信息，对每个所述有效目标车辆进行分组，得到分组信息。
在一些实施例中，分组模块501，还用于在所述得到分组信息后，基于历史交通流信息修正所述分组信息；相应地，确定模块502，用于基于所述历史交通流信息、每个所述有效目标车辆的状态信息和修正后的分组信息，确定每个所述有效目标车辆的拟合权重。
在一些实施例中，分组模块501，针对每个所述有效目标车辆：基于该有效目标车辆的状态信息和所述本车的行驶轨迹，确定该有效目标车辆所在的相对车道；基于所述相对车道，确定该有效目标车辆的分组信息。
在一些实施例中，所述分组信息包括以下五组：本车所在车道的左侧第一车道分组、本车所在车道的左侧第二车道分组、本车所在车道的右侧第一车道分组、本车所在车道的右侧第二车道分组和其他分组。
在一些实施例中，拟合模块503，还用于在拟合生成一条或多条当前交通流信息之前，判断目标车辆数量是否满足预设的拟合数量；若满足，则判断本车的行驶状态是否为直行；若本车的行驶状态为直行，则获取缓存的历史车辆；更新每个所述历史车辆的坐标，得到每个历史车辆对应的虚拟车辆 及每个所述虚拟车辆的状态信息；基于历史交通流信息、所述本车的运动信息、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的分组信息，确定每个所述虚拟车辆的拟合权重；
相应地，拟合模块503，用于基于每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的拟合权重，拟合生成一条或多条当前交通流信息。
在一些实施例中，所述历史车辆满足缓存条件；其中，所述缓存条件用于筛选与本车同向行驶的车辆且该车辆不属于其他分组。
在一些实施例中，所述缓存条件包括：车辆位于本车前方且与本车的相对距离大于预设的相对距离阈值；车辆相对于本车的速度的方向与本车的行驶方向之间的夹角小于预设的夹角阈值；车辆相对于本车的侧向距离处于预设的侧向距离范围内。
在一些实施例中，所述缓存条件还包括：车辆的生命周期大于预设的生命周期阈值。
在一些实施例中，确定模块502针对每个所述目标车辆：基于该目标车辆的状态信息和该目标车辆的分组信息，确定该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重；将该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重进行相乘，得到该目标车辆的拟合权重。
在一些实施例中，确定模块502针对每个所述虚拟车辆：基于该虚拟车辆的状态信息和该虚拟车辆的分组信息，确定该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重；将该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重进行相乘，得到该虚拟车辆的拟合权重。
在一些实施例中，每个分组对应一条交通流；拟合模块503针对一个分组：基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式；基于选择的拟合方式，利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息。
在一些实施例中，拟合模块503基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式包括：判断该分组中目标车辆的数量是否满足第一数量条件，且纵向分布距离是否大于预设的第一距离阈值；若满足第一数量条件，且大于预设的第一距离阈值，则选择一阶拟合和二阶拟合；否则，判断该分组中目标车辆的数量是否满足第二数量条件，且纵向分布距离是否大于预设的第二距离阈值；若满足第二数量条件，且大于预设的第二距离阈值，则选择一阶拟合；否则，不进行拟合。
在一些实施例中，若选择一阶拟合和二阶拟合，则拟合模块503拟合生成该分组对应的当前交通流信息包括：拟合生成该分组对应的一阶拟合结果，并确定所述一阶拟合结果的第一均方差；拟合生成该分组对应的二阶拟合结果，并确定所述二阶拟合结果的第二均方差；比较所述第一均方差和所述第二均方差，若所述第一均方差大于所述第二均方差的预设倍数，则选择所述二阶拟合结果为该分组 对应的当前交通流信息；否则，选择所述一阶拟合结果为该分组对应的当前交通流信息。
在一些实施例中，拟合模块503利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息包括：基于该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的初始交通流信息；基于历史交通流信息和所述本车的运动信息，对所述初始交通流信息进行约束，得到该分组对应的当前交通流信息。
在一些实施例中，所述约束包括：位置约束、航向约束和/或曲率约束。
在一些实施例中，所述位置约束包括：基于所述历史交通流信息对应的置信度约束所述初始交通流信息对应的拟合参数中的常数项的变化量；所述常数项的变化量与所述历史交通流信息对应的置信度之间为反相关；所述常数项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
在一些实施例中，所述航向约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的一次项的变化量；其中，所述一次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述一次项的变化量与所述本车的转向幅度之间为正相关，和/或，所述一次项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
在一些实施例中，所述曲率约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的二次项的变化量；其中，所述二次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述二次项的变化量与所述本车的速度之间为反相关，和/或，所述二次项的变化量与所述本车的方向盘转速之间为正相关。
在一些实施例中，所述交通流信息的确定装置还包括图5中未示出的置信度确定单元，用于：基于所述当前交通流信息，确定置信度增量；基于历史交通流信息的置信度和所述置信度增量，确定所述当前交通流信息对应的置信度；其中，所述置信度增量与所述当前交通流信息对应的拟合结果均方差之间为反相关，所述置信度增量与所述当前交通流信息对应的拟合数量之间为正相关，和/或，所述置信度增量与所述当前交通流信息对应的目标车辆的纵向分布距离之间为正相关。
置信度确定单元基于所述当前交通流信息，确定置信度增量包括：若两条当前交通流信息存在交叉，则比较两条交通流对应的拟合数量和纵向分布距离；确定拟合数量较少或纵向分布距离较短的当前交通流信息对应的置信度增量为负。
在一些实施例中，所述交通流信息的确定装置还包括图5中未示出的参考路径确定单元，用于基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定本车行驶参考路径；所述确定本车行驶参考路径包括：若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定本车行驶参考路径的参数为本车两侧的当前交通流信息对应的拟合参数的加权平均值；其中，本车两侧的当前交通流信息对应的拟合参数的权重与拟合参数的变化率之间为反相关；若本车仅一侧的当前交通流信息对应的置信度高于预设置信度门限，则确定本车行驶参考路径的参数中一次项和二次项与该侧当前交通流信息对应的拟合参数中一次项和二次项相同，本车行驶参考路径的参数中常数项为零；若本车两侧的当前交通流信息对应的置信度均低于预设置信度门限，则确定本车行驶参考路径的参数均为零。
在一些实施例中，所述交通流信息的确定装置还包括图5中未示出的辅助定位标志确定单元，用于基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定辅助定位标志；所述确定辅助定位标志包括：若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定辅助定位标志为第一标志；若本车左侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第二标志；若本车右侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第三标志。
在一些实施例中，交通流信息的确定装置中各单元的划分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，例如分组模块501、确定模块502和拟合模块503中的至少两个单元可以实现为一个单元；分组模块501、确定模块502或拟合模块503也可以划分为多个子单元。可以理解的是，各个单元或子单元能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。本领域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。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
本领域的技术人员能够理解，尽管在此所述的一些实施例包括其它实施例中所包括的某些特征而不是其它特征，但是不同实施例的特征的组合意味着处于本公开的范围之内并且形成不同的实施例。
本领域的技术人员能够理解，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
虽然结合附图描述了本公开的实施方式，但是本领域技术人员可以在不脱离本公开的精神和范围的情况下做出各种修改和变型，这样的修改和变型均落入由所附权利要求所限定的范围之内。
工业实用性
本公开的至少一个实施例中，通过对多个目标车辆进行分组，并确定每个目标车辆的拟合权重，进而可基于分组信息、拟合权重、本车的运动信息和目标车辆的状态信息，拟合生成当前交通流信息，实现对交通流的感知。本公开实施例可基于当前交通流信息和当前交通流信息对应的置信度，确定辅助定位标志，辅助定位标志独立于车道线信息，可在部分多车道场景，车道线不足以确定唯一车道时，辅助确定本车所在车道，为L3级自动驾驶系统的远距离路径规划提供了必要依据。另外，本公开实施例可基于当前交通流信息进行本车横向控制，摆脱了对车道线信息的依赖，有利于在拥堵环境下维持本车的横向控制。具有工业实用性。

Claims (26)

  1. 一种交通流信息的确定方法，其特征在于，包括：
    基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息；
    基于每个所述目标车辆的状态信息和所述分组信息，确定每个所述目标车辆的拟合权重；
    基于所述本车的运动信息、每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重和所述分组信息，拟合生成一条或多条当前交通流信息。
  2. 根据权利要求1所述的方法，其特征在于，所述基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息包括：
    基于所述本车的运动信息，确定本车的行驶轨迹；
    基于所述本车的运动信息和每个所述目标车辆的状态信息，筛选一个或多个有效目标车辆；
    基于所述本车的行驶轨迹和每个所述有效目标车辆的状态信息，对每个所述有效目标车辆进行分组，得到分组信息。
  3. 根据权利要求2所述的方法，其特征在于，所述得到分组信息后，所述方法还包括：
    基于历史交通流信息修正所述分组信息；
    相应地，基于所述历史交通流信息、每个所述有效目标车辆的状态信息和修正后的分组信息，确定每个所述有效目标车辆的拟合权重。
  4. 根据权利要求2所述的方法，其特征在于，所述基于所述本车的行驶轨迹和每个所述有效目标车辆的状态信息，对每个所述有效目标车辆进行分组，得到分组信息包括：
    针对每个所述有效目标车辆：
    基于该有效目标车辆的状态信息和所述本车的行驶轨迹，确定该有效目标车辆所在的相对车道；
    基于所述相对车道，确定该有效目标车辆的分组信息。
  5. 根据权利要求1至4任一项所述的方法，其特征在于，所述分组信息包括以下五组：本车所在车道的左侧第一车道分组、本车所在车道的左侧第二车道分组、本车所在车道的右侧第一车道分组、本车所在车道的右侧第二车道分组和其他分组。
  6. 根据权利要求5所述的方法，其特征在于，所述拟合生成一条或多条当前交通流信息之前，所述方法还包括：
    判断目标车辆数量是否满足预设的拟合数量；若满足，则判断本车的行驶状态是否为直行；
    若本车的行驶状态为直行，则获取缓存的历史车辆；
    更新每个所述历史车辆的坐标，得到每个历史车辆对应的虚拟车辆及每个所述虚拟车辆的状态信息；
    基于历史交通流信息、所述本车的运动信息、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的分组信息，确定每个所述虚拟车辆的拟合权重；
    相应地，所述拟合生成一条或多条当前交通流信息包括：基于每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重、每个所述虚拟车辆的状态信息和每个所述虚拟车辆的拟合权重，拟合生 成一条或多条当前交通流信息。
  7. 根据权利要求6所述的方法，其特征在于，所述历史车辆满足缓存条件；其中，所述缓存条件用于筛选与本车同向行驶的车辆且该车辆不属于其他分组。
  8. 根据权利要求7所述的方法，其特征在于，所述缓存条件包括：
    车辆位于本车前方且与本车的相对距离大于预设的相对距离阈值；
    车辆相对于本车的速度的方向与本车的行驶方向之间的夹角小于预设的夹角阈值；
    车辆相对于本车的侧向距离处于预设的侧向距离范围内。
  9. 根据权利要求8所述的方法，其特征在于，所述缓存条件还包括：车辆的生命周期大于预设的生命周期阈值。
  10. 根据权利要求1所述的方法，其特征在于，所述确定每个所述目标车辆的拟合权重包括：
    针对每个所述目标车辆：
    基于该目标车辆的状态信息和该目标车辆的分组信息，确定该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重；
    将该目标车辆的侧向速度偏离权重、该目标车辆的侧向位移偏离权重、该目标车辆的生命周期权重和该目标车辆的速度权重进行相乘，得到该目标车辆的拟合权重。
  11. 根据权利要求6所述的方法，其特征在于，所述确定每个所述虚拟车辆的拟合权重包括：
    针对每个所述虚拟车辆：
    基于该虚拟车辆的状态信息和该虚拟车辆的分组信息，确定该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重；
    将该虚拟车辆的侧向速度偏离权重、该虚拟车辆的侧向位移偏离权重、该虚拟车辆的生命周期权重、该虚拟车辆的速度权重和该虚拟车辆的权重进行相乘，得到该虚拟车辆的拟合权重。
  12. 根据权利要求1所述的方法，其特征在于，每个分组对应一条交通流；所述拟合生成一条或多条当前交通流信息包括：
    针对一个分组：
    基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式；
    基于选择的拟合方式，利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息。
  13. 根据权利要求12所述的方法，其特征在于，所述基于该分组中目标车辆的数量以及该分组中所有目标车辆的纵向分布距离，选择拟合方式包括：
    判断该分组中目标车辆的数量是否满足第一数量条件，且纵向分布距离是否大于预设的第一距离阈值；
    若满足第一数量条件，且大于预设的第一距离阈值，则选择一阶拟合和二阶拟合；否则，判断该分组中目标车辆的数量是否满足第二数量条件，且纵向分布距离是否大于预设的第二距离阈值；
    若满足第二数量条件，且大于预设的第二距离阈值，则选择一阶拟合；否则，不进行拟合。
  14. 根据权利要求13所述的方法，其特征在于，若选择一阶拟合和二阶拟合，则所述拟合生成该分组对应的当前交通流信息包括：
    拟合生成该分组对应的一阶拟合结果，并确定所述一阶拟合结果的第一均方差；
    拟合生成该分组对应的二阶拟合结果，并确定所述二阶拟合结果的第二均方差；
    比较所述第一均方差和所述第二均方差，若所述第一均方差大于所述第二均方差的预设倍数，则选择所述二阶拟合结果为该分组对应的当前交通流信息；否则，选择所述一阶拟合结果为该分组对应的当前交通流信息。
  15. 根据权利要求12所述的方法，其特征在于，所述利用所述本车的运动信息、该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的当前交通流信息包括：
    基于该分组中每个目标车辆的状态信息和该分组中每个目标车辆的拟合权重，拟合生成该分组对应的初始交通流信息；
    基于历史交通流信息和所述本车的运动信息，对所述初始交通流信息进行约束，得到该分组对应的当前交通流信息。
  16. 根据权利要求15所述的方法，其特征在于，所述约束包括：位置约束、航向约束和/或曲率约束。
  17. 根据权利要求16所述的方法，其特征在于，所述位置约束包括：基于所述历史交通流信息对应的置信度约束所述初始交通流信息对应的拟合参数中的常数项的变化量；
    所述常数项的变化量与所述历史交通流信息对应的置信度之间为反相关；所述常数项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
  18. 根据权利要求16所述的方法，其特征在于，所述航向约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的一次项的变化量；其中，所述一次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述一次项的变化量与所述本车的转向幅度之间为正相关，和/或，所述一次项的变化量与所述初始交通流信息对应的拟合数量之间为正相关。
  19. 根据权利要求16所述的方法，其特征在于，所述曲率约束包括：基于所述历史交通流信息对应的置信度和/或所述本车的运动信息，约束所述初始交通流信息对应的拟合参数中的二次项的变化量；其中，所述二次项的变化量与所述历史交通流信息对应的置信度之间为反相关，所述二次项的变化量与所述本车的速度之间为反相关，和/或，所述二次项的变化量与所述本车的方向盘转速之间为正相关。
  20. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    基于所述当前交通流信息，确定置信度增量；
    基于历史交通流信息的置信度和所述置信度增量，确定所述当前交通流信息对应的置信度；
    其中，所述置信度增量与所述当前交通流信息对应的拟合结果均方差之间为反相关，所述置信度增量与所述当前交通流信息对应的拟合数量之间为正相关，和/或，所述置信度增量与所述当前交通 流信息对应的目标车辆的纵向分布距离之间为正相关。
  21. 根据权利要求20所述的方法，其特征在于，所述基于所述当前交通流信息，确定置信度增量包括：若两条当前交通流信息存在交叉，则比较两条交通流对应的拟合数量和纵向分布距离；确定拟合数量较少或纵向分布距离较短的当前交通流信息对应的置信度增量为负。
  22. 根据权利要求20所述的方法，其特征在于，所述方法还包括：基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定本车行驶参考路径；
    所述确定本车行驶参考路径包括：
    若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定本车行驶参考路径的参数为本车两侧的当前交通流信息对应的拟合参数的加权平均值；其中，本车两侧的当前交通流信息对应的拟合参数的权重与拟合参数的变化率之间为反相关；
    若本车仅一侧的当前交通流信息对应的置信度高于预设置信度门限，则确定本车行驶参考路径的参数中一次项和二次项与该侧当前交通流信息对应的拟合参数中一次项和二次项相同，本车行驶参考路径的参数中常数项为零；
    若本车两侧的当前交通流信息对应的置信度均低于预设置信度门限，则确定本车行驶参考路径的参数均为零。
  23. 根据权利要求20所述的方法，其特征在于，所述方法还包括：基于所述当前交通流信息和所述当前交通流信息对应的置信度，确定辅助定位标志；
    所述确定辅助定位标志包括：
    若本车两侧的当前交通流信息对应的置信度均高于预设置信度门限，则确定辅助定位标志为第一标志；若本车左侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第二标志；若本车右侧的当前交通流信息对应的置信度高于预设置信度门限，则确定辅助定位标志为第三标志。
  24. 一种交通流信息的确定装置，其特征在于，包括：
    分组模块，用于基于本车的运动信息和一个或多个目标车辆的状态信息，对每个所述目标车辆进行分组，得到分组信息；
    确定模块，用于基于每个所述目标车辆的状态信息和所述分组信息，确定每个所述目标车辆的拟合权重；
    拟合模块，用于基于所述本车的运动信息、每个所述目标车辆的状态信息、每个所述目标车辆的拟合权重和所述分组信息，拟合生成一条或多条当前交通流信息。
  25. 一种电子设备，其特征在于，包括：处理器和存储器；所述处理器通过调用所述存储器存储的程序或指令，用于执行如权利要求1至23任一项所述方法的步骤。
  26. 一种非暂态计算机可读存储介质，其特征在于，所述非暂态计算机可读存储介质存储程序或指令，所述程序或指令使计算机执行如权利要求1至23任一项所述方法的步骤。
PCT/CN2020/127499 2020-11-09 2020-11-09 一种交通流信息的确定方法、装置、电子设备和存储介质 Ceased WO2022095023A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
CN202080002861.7A CN112567439B (zh) 2020-11-09 2020-11-09 一种交通流信息的确定方法、装置、电子设备和存储介质
JP2023527374A JP7580155B2 (ja) 2020-11-09 2020-11-09 交通流情報の決定方法、装置、電子機器及び記憶媒体
US18/035,936 US12469384B2 (en) 2020-11-09 2020-11-09 Method and device for determining traffic stream information, electronic equipment and storage medium
KR1020237018513A KR102620325B1 (ko) 2020-11-09 2020-11-09 교통 흐름 정보의 결정 방법, 장치, 전자 기기 및 저장 매체
PCT/CN2020/127499 WO2022095023A1 (zh) 2020-11-09 2020-11-09 一种交通流信息的确定方法、装置、电子设备和存储介质
EP20960481.8A EP4242998B1 (en) 2020-11-09 2020-11-09 Traffic stream information determination method and apparatus, electronic device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2020/127499 WO2022095023A1 (zh) 2020-11-09 2020-11-09 一种交通流信息的确定方法、装置、电子设备和存储介质

Publications (1)

Publication Number Publication Date
WO2022095023A1 true WO2022095023A1 (zh) 2022-05-12

Family

ID=75034957

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/127499 Ceased WO2022095023A1 (zh) 2020-11-09 2020-11-09 一种交通流信息的确定方法、装置、电子设备和存储介质

Country Status (6)

Country Link
US (1) US12469384B2 (zh)
EP (1) EP4242998B1 (zh)
JP (1) JP7580155B2 (zh)
KR (1) KR102620325B1 (zh)
CN (1) CN112567439B (zh)
WO (1) WO2022095023A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115294776A (zh) * 2022-06-23 2022-11-04 北京北大千方科技有限公司 基于时间切片统计车辆通行量的方法、装置、设备及介质
CN115468778A (zh) * 2022-09-14 2022-12-13 北京百度网讯科技有限公司 车辆测试方法、装置、电子设备及存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11995983B2 (en) * 2020-10-16 2024-05-28 Grabtaxi Holdings Pte. Ltd. Method, electronic device, and system for detecting overspeeding
EP4242998B1 (en) * 2020-11-09 2025-02-19 UISEE (Shanghai) Automotive Technologies Ltd Traffic stream information determination method and apparatus, electronic device and storage medium
CN114005279A (zh) * 2021-11-11 2022-02-01 蔚来汽车科技(安徽)有限公司 车端交通流预测方法、装置、存储介质和车辆
CN115019554B (zh) * 2022-05-30 2024-03-26 中国第一汽车股份有限公司 一种车辆告警方法、装置、电子设备及存储介质
CN115027470B (zh) * 2022-06-09 2025-09-09 中国第一汽车股份有限公司 车辆纵向控制方法、装置以及车辆
CN115879294B (zh) * 2022-11-30 2024-04-26 西部科学城智能网联汽车创新中心(重庆)有限公司 一种基于多车环境感知的全样车流轨迹生成方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110109475A1 (en) * 2009-11-12 2011-05-12 Gm Global Technology Operations, Inc. Travel Lane Advisor
CN103903019A (zh) * 2014-04-11 2014-07-02 北京工业大学 一种多车道车辆轨迹时空图的自动生成方法
CN108171967A (zh) * 2017-11-24 2018-06-15 华为技术有限公司 一种交通控制方法及装置
US20190355258A1 (en) * 2017-02-09 2019-11-21 Mitsubishi Electric Corporation Information processing device, information processing method, and computer readable medium
CN111738207A (zh) * 2020-07-13 2020-10-02 腾讯科技（深圳）有限公司 车道线检测方法、装置、电子设备及可读存储介质

Family Cites Families (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000132783A (ja) * 1998-10-26 2000-05-12 Semba Corp 交通流シミュレータ、環境解析システム、交通流のシミュレーション方法および記憶媒体
JP2003156413A (ja) * 2001-11-20 2003-05-30 Toyo Tire & Rubber Co Ltd 車体振動のシミュレーション方法
JP5185663B2 (ja) 2008-03-12 2013-04-17 富士重工業株式会社 車両用運転支援装置
JP5868586B2 (ja) * 2009-11-04 2016-02-24 ヴァレオ・シャルター・ウント・ゼンゾーレン・ゲーエムベーハー ビデオ画像に基づく道路特性分析、車線検出、および車線逸脱防止方法および装置
KR101526816B1 (ko) * 2014-09-12 2015-06-05 현대자동차주식회사 차선 추정 시스템 및 그 방법
JP5984155B2 (ja) * 2014-09-26 2016-09-06 インターナショナル・ビジネス・マシーンズ・コーポレーションＩｎｔｅｒｎａｔｉｏｎａｌ Ｂｕｓｉｎｅｓｓ Ｍａｃｈｉｎｅｓ Ｃｏｒｐｏｒａｔｉｏｎ 情報処理装置、プログラム、及び、情報処理方法
KR102215325B1 (ko) 2017-02-28 2021-02-15 현대자동차주식회사 차량의 위치 추정 장치 및 방법과 이를 이용한 차량
CN107402131B (zh) * 2017-08-03 2018-05-25 安徽大学 一种基于轨边声学信号时频脊线的高速列车运动参数识别方法
US10761535B2 (en) 2018-08-21 2020-09-01 GM Global Technology Operations LLC Intelligent vehicle navigation systems, methods, and control logic for multi-lane separation and trajectory extraction of roadway segments
CN109242912A (zh) 2018-08-29 2019-01-18 杭州迦智科技有限公司 采集装置外参标定方法、电子设备、存储介质
CN109541583B (zh) * 2018-11-15 2020-05-01 众安信息技术服务有限公司 一种前车距离检测方法及系统
CN109683617B (zh) * 2018-12-27 2022-06-28 潍柴动力股份有限公司 一种自动驾驶方法、装置及电子设备
CN111176269B (zh) * 2019-10-11 2022-07-05 中国第一汽车股份有限公司 车辆的航向调整方法、装置、车辆及存储介质
CN110703754B (zh) * 2019-10-17 2021-07-09 南京航空航天大学 一种自动驾驶车辆路径与速度高度耦合的轨迹规划方法
CN110782667A (zh) 2019-10-30 2020-02-11 北京百度网讯科技有限公司 信号灯分时段配时方法、装置、电子设备及存储介质
CN111241224B (zh) * 2020-01-10 2021-05-11 福瑞泰克智能系统有限公司 目标距离估计的方法、系统、计算机设备和存储介质
CN111325187B (zh) * 2020-03-23 2023-10-20 北京经纬恒润科技股份有限公司 一种车道位置的识别方法及装置
EP4242998B1 (en) * 2020-11-09 2025-02-19 UISEE (Shanghai) Automotive Technologies Ltd Traffic stream information determination method and apparatus, electronic device and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110109475A1 (en) * 2009-11-12 2011-05-12 Gm Global Technology Operations, Inc. Travel Lane Advisor
CN103903019A (zh) * 2014-04-11 2014-07-02 北京工业大学 一种多车道车辆轨迹时空图的自动生成方法
US20190355258A1 (en) * 2017-02-09 2019-11-21 Mitsubishi Electric Corporation Information processing device, information processing method, and computer readable medium
CN108171967A (zh) * 2017-11-24 2018-06-15 华为技术有限公司 一种交通控制方法及装置
CN111738207A (zh) * 2020-07-13 2020-10-02 腾讯科技（深圳）有限公司 车道线检测方法、装置、电子设备及可读存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4242998A4 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115294776A (zh) * 2022-06-23 2022-11-04 北京北大千方科技有限公司 基于时间切片统计车辆通行量的方法、装置、设备及介质
CN115294776B (zh) * 2022-06-23 2024-04-12 北京北大千方科技有限公司 基于时间切片统计车辆通行量的方法、装置、设备及介质
CN115468778A (zh) * 2022-09-14 2022-12-13 北京百度网讯科技有限公司 车辆测试方法、装置、电子设备及存储介质
CN115468778B (zh) * 2022-09-14 2023-08-15 北京百度网讯科技有限公司 车辆测试方法、装置、电子设备及存储介质

Also Published As

Publication number Publication date
EP4242998B1 (en) 2025-02-19
JP7580155B2 (ja) 2024-11-11
KR102620325B1 (ko) 2024-01-03
CN112567439A (zh) 2021-03-26
CN112567439B (zh) 2022-11-29
EP4242998A4 (en) 2023-11-22
US12469384B2 (en) 2025-11-11
KR20230098633A (ko) 2023-07-04
EP4242998A1 (en) 2023-09-13
JP2023548879A (ja) 2023-11-21
US20230419824A1 (en) 2023-12-28

Similar Documents

Publication Publication Date Title
CN112567439B (zh) 一种交通流信息的确定方法、装置、电子设备和存储介质
EP3517893B1 (en) Path and speed optimization fallback mechanism for autonomous vehicles
CN105292116B (zh) 自动驾驶车辆的车道变换路径规划算法
US20220284619A1 (en) Offline optimization of sensor data for agent trajectories
US20260022942A1 (en) Systems and methods for deriving path-prior data using collected trajectories
US9283967B2 (en) Accurate curvature estimation algorithm for path planning of autonomous driving vehicle
KR20210032283A (ko) Radar를 사용한 확장 대상체 추적
CN112612287A (zh) 一种自动驾驶汽车局部路径规划系统、方法、介质及设备
CN112394725B (zh) 用于自动驾驶的基于预测和反应视场的计划
CN111738207A (zh) 车道线检测方法、装置、电子设备及可读存储介质
CN108061555B (zh) 一种车辆定位误差矫正方法及装置
US12169949B2 (en) Camera to camera calibration
JP2018192954A (ja) 運転支援装置
CN116118770B (zh) 面向稳健自动驾驶控制的车辆感知系统的自适应合理化器
WO2023092451A1 (zh) 预测可行驶车道的方法和装置
CN113227831A (zh) 基于多传感器数据融合的护栏估计方法和车载设备
US20250010880A1 (en) Lateral controller for autonomous vehicles
JP7149082B2 (ja) 走行支援装置の走行支援方法及び走行支援装置
CN114355333A (zh) 一种自动驾驶基于控制信息的目标筛选方法及系统、车辆
JP7571742B2 (ja) 車両制御装置、車両制御用コンピュータプログラム及び車両制御方法
JP2022178813A (ja) 車両運転支援システム及び車両運転支援方法
CN116529142A (zh) 适应于感知的控制系统和方法
CN115303288A (zh) 一种车辆控制方法、控制装置及摄像头装置
JP7129269B2 (ja) 走行履歴記憶方法及び走行履歴記憶装置
JP7856050B2 (ja) 車両における自動運転機能の管理方法及び管理システム

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: 20960481

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2023527374

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 18035936

Country of ref document: US

ENP Entry into the national phase

Ref document number: 20237018513

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020960481

Country of ref document: EP

Effective date: 20230609

WWG Wipo information: grant in national office

Ref document number: 18035936

Country of ref document: US