EP4154166A1 - Verfahren, computer-implementiertes werkzeug, fahrzeug-steuerungseinheit und fahrzeug zum verorten von hindernisobjekten in landmarkengekennzeichneten fahrstreckenbereichen von fahrzeugen - Google Patents
Verfahren, computer-implementiertes werkzeug, fahrzeug-steuerungseinheit und fahrzeug zum verorten von hindernisobjekten in landmarkengekennzeichneten fahrstreckenbereichen von fahrzeugenInfo
- Publication number
- EP4154166A1 EP4154166A1 EP21746412.2A EP21746412A EP4154166A1 EP 4154166 A1 EP4154166 A1 EP 4154166A1 EP 21746412 A EP21746412 A EP 21746412A EP 4154166 A1 EP4154166 A1 EP 4154166A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- sensor measurement
- landmark
- sensor
- measurement data
- vehicle
- 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.)
- Pending
Links
Classifications
-
- 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
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- 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
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
- B60W40/06—Road conditions
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- 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
- B60W2420/00—Indexing codes relating to the type of sensors based on the principle of their operation
- B60W2420/40—Photo, light or radio wave sensitive means, e.g. infrared sensors
- B60W2420/403—Image sensing, e.g. optical camera
Definitions
- the invention relates to a method for locating obstacle objects in landmark-marked route areas of vehicles according to the preamble ff of patent claim 1, a computer-implemented tool for locating obstacle objects in landmark-marked route areas of vehicles according to the preamble ff of patent claim 9, a vehicle control unit for locating obstacle objects in landmark-marked route areas of vehicles according to the preamble ff of patent claim 16 and a vehicle for locating obstacle objects in landmark-marked route areas of vehicles according to the preamble ff of patent claim 17 .
- autonomous driving When driving vehicles as a mobile means of transport for transporting people, goods, tools or machines - regardless of whether the driving is assisted, partially automated, highly automated, fully automated or autonomous (one then speaks collectively of "autonomous driving"), it is a fundamental one Function of such autonomous vehicles to avoid collisions with obstacle objects.
- Such vehicles travel on land (as land vehicles), on water (as water vehicles) and/or in the air (as aircraft).
- land vehicles land vehicles
- water vehicles water vehicles
- air air
- only those vehicles of the type defined above are considered that move or are moved relative to routes marked with landmarks.
- a landmark is a topographical object that is usually visible from afar for spatial orientation and terrestrial Navigation, which may be marked on maps by special map symbols. Because it is stored in the map, it is therefore a specific feature or topological object in the world that can be recognized by a vehicle, so that an association can be established between sensory recognition and the map.
- the landmark may be a commonly occurring object such as a lamp post, a specially placed object such as reflective markers and/or QR ⁇ Quick Response> codes, or it may be a recognizable one resulting from recognition Patterns such as "Scale Invariant Feature Transform ⁇ SIFT>" features, cf. https://en.wikipedia.org/wiki/Scale-invariant feature transform.
- a sensory detection of an obstacle object is necessary for collision avoidance and the detected obstacle object must be assigned to a critical area.
- the critical area usually includes a lane or a route of the vehicle itself and possibly an area surrounding the lane or the route.
- the critical area results from a combination of map material, ego localization or self-localization and planned route or route.
- the present invention also speaks of a lane or route area with regard to collision avoidance and the sensory detection of the obstacle objects.
- the assignment or localization of the obstacle object to a critical area is disturbed by various influencing factors. These are, for example: faulty extrinsic sensor calibration, faulty ego localization (self-localization), faulty map material, weaving on the track in rail vehicles, low horizontal and vertical angular resolution of the sensors, which is particularly relevant for large ranges.
- a probabilistic fusion of data from multiple sensors is a way to reduce positional inaccuracies. In this way, the disadvantages of some sensors can be compensated for by the advantages of other sensors.
- the vehicle locates itself.
- Various information can be used, e.g. B. GPS information “Global Positioning System ⁇ GPS>”, sensor information (e.g. radar ⁇ Radio Detection and Ranging>, lidar ⁇ Light Detection and Ranging>, camera) and/or also landmark information from the map material.
- GPS information “Global Positioning System ⁇ GPS>”
- sensor information e.g. radar ⁇ Radio Detection and Ranging>, lidar ⁇ Light Detection and Ranging>, camera
- landmark information from the map material.
- z. B. NVIDIA-Drive ( v g 1 . https : / /developer . nvidia . com/ drive /drive -percept ion ) known obstacle localization, where a neural network , e.g. B. DriveNet ; Deconvolutional neural network (DNN), for which obstacle object recognition is used and an independent neural network for lane and landmark recognition, e.g. B. MapNet ; Deconvolutional Neural Network (DNN), the landmarks in step 2 are used for localization
- a neural network e.g. B. DriveNet
- DNN Deconvolutional neural network
- MapNet Deconvolutional Neural Network
- the object on which the invention is based is to specify a method, a computer-implemented tool, a vehicle control unit and a vehicle for locating obstacle objects in landmark-marked route areas of vehicles, without locating the obstacle objects precisely in a known manner, nor one for this purpose having to strive for the required coordinate transformation.
- this object is achieved by the features specified in the characterizing part of patent claim 16.
- this object relating to a vehicle is achieved by the features specified in patent claim 17 .
- an obstacle object in a landmark-marked route area of a vehicle is then determined by (i) Associating sensor measurement data, recorded by sensors of a vehicle sensor system and belonging to sensor measurement objects, through which a landmark and the obstacle object can be represented, with stored landmark reference data, in particular stored in the form of map material, and (ii) determining a sensor detection-specific data Localization distance between a sensor measurement object with unassociated sensor measurement data and a landmark with associated sensor measurement data on the basis of the associated and unassociated sensor measurement data by sensor acquisition-specific information contained in this sensor measurement data in relation to g are placed in relation to each other on the localization distances to be determined, located on a route of the route area where it represents an obstacle for the vehicle, or its position is determined if either on the basis of the determined sensor detection-specific localization distance or on the basis the determined sensor detection-specific localization distance and using the landmark reference
- the localization of the vehicle in relation to the route area is not to be determined with the help of landmarks, but rather the position of a potentially dangerous obstacle object with the help of the sensors of the vehicle sensor system with the aid of Acquisition of landmark information is determined by relations between known landmarks and detected th Obstacles ects in relation to the route of the vehicle in the route area as a critical area produced.
- the teaching according to the invention does not primarily aim to locate the obstacle object that cannot be assigned to the route area very precisely and absolutely to the vehicle, but rather to assess the risk of whether this obstacle object is due to whether or not its location in a critical area, such as the driving route, constitutes a potential obstacle to the vehicle.
- the improved and closer connection between map, vehicle, obstacle objects and critical areas results in increased robustness with regard to the obstacle object localization and assignment to critical areas. This will also The number of false positives is reduced, which in turn increases overall system reliability.
- landmarks Due to the use of known landmarks and potentially also landmark formations, obstacle objects can be assigned to areas without having to precisely localize them individually.
- the landmarks serve as anchors that can greatly reduce the uncertainty in the object detection and localization problem.
- the teaching according to the invention can be used in many different ways, e.g. B. in applications involving autonomous vehicles. But there is also application potential in intelligent traffic engineering and "Automated Guided Vehicles ⁇ AGV's>".
- FIGURE 1 shows a vehicle scenario for locating obstacle objects in a route area marked with landmarks
- FIGURE 2 shows a modified vehicle scenario for locating obstacles in a route area marked by landmarks with an obstacle object extending vertically to a route of the route area
- FIGURE 3 shows a modified vehicle scenario for locating obstacles in a route area marked by landmarks with a route in which the landmarks are arranged in a grouping or formation
- FIG. 4 shows another vehicle scenario for locating an obstacle based on the scenario according to FIG 3 with a detection of landmarks in the route area at different localization times.
- FIG. 1 shows a vehicle scenario for locating obstacle objects HO in a route area FSB marked with landmarks.
- the route range FSB includes a route FS for a vehicle FZ, which as a land vehicle and in this respect z.
- B. designed as a passenger car ⁇ car> is traveling in a vehicle lane FZSP of a road STR in the direction of an arrow drawn in on the vehicle FZ.
- the route area FSB also includes (a) at least one landmark LM arranged along the route FS with the definition given above in the introduction to the description, which is also suitable for using modeled, situational relationships in the route area, (b) at least one of the at least one additional landmark LM, reference object RO with a defined, constant distance from the route FS of the vehicle route area FSB, the z. B. a noise protection wall LSW (2-dimensional feature of the route area) on the road STR, and (c) at least one obstacle HO arranged along the route FS and/or on the route FS, which poses a danger to the vehicle FZ , should it be on the route FS .
- LSW noise protection wall
- Further reference objects can be features that already exist in the vicinity of the route area FSB or features that are placed in a targeted manner. Examples can be mentioned:
- sensor-specific markers such as e.g. B. Reflectors for lidar and radar, specific patterns for cameras. Characteristics resulting automatically from the algorithm, e.g. B. SI FT features.
- the vehicle FZ contains a vehicle sensor system FZS with at least one sensor SS and a vehicle control unit FZSE, which are used for sensor control and sensor data transmission are connected .
- the vehicle sensor system FZS contains two sensors SS of different modality with regard to the obstacle object location according to the invention, which is explained in more detail below, to increase the robustness and to reduce the uncertainty in the obstacle object location.
- the vehicle sensor system FZS contains a sensor SS, which is based on a Cartesian coordinate system SS KKS, a spherical coordinate system SS SKS or a cylindrical coordinate system SS ZK s , and another sensor SS, which is based on an imaging principle SS B GP .
- the sensor SS which is based on the Cartesian coordinate system SS KKS , the spherical coordinate system SS SKS or the cylindrical coordinate system SS ZK s , is preferably a radar sensor based on a "Radio Detection and Ranging ⁇ Radar>” principle or a "Light Detection and Ranging "Lidar>” principle based lidar sensor used, while as a sensor SS, which is based on an imaging principle SS B GP, preferably a camera sensor is used.
- the vehicle control unit FZSE in the vehicle FZ contains - according to an option "A" shown in FIG. for locating the obstacle object, a storage device SPE for storing (i) landmark reference data LM-RD, which represent the at least one landmark LM and which are stored in the storage device SPE, for example as map material KM, and (ii) reference object data ROD, which represent the at least one reference object RO and which are also stored in the memory device SPE, for example as map material KM, and a control device STE, which is connected to the memory device SPE.
- a storage device SPE for storing (i) landmark reference data LM-RD, which represent the at least one landmark LM and which are stored in the storage device SPE, for example as map material KM, and (ii) reference object data ROD, which represent the at least one reference object RO and which are also stored in the memory device SPE, for example as map material KM
- a control device STE which is connected to the memory device SPE.
- a computer-implemented tool CIW can now be uploaded to this control device STE for carrying out the obstacle object localization in the vehicle control unit FZSE, which is preferably designed as a computer program product and is referred to as application software ⁇ APP>.
- the computer-implemented tool CIW has a non-volatile, readable memory SP, in which processor-readable control program commands of a program module PGM for locating obstacle objects are stored, and one with the memory for the obstacle object j ect location with the upload to the control device STE SP-connected processor PZ, which executes the control program commands of the program module PGM for obstacle object j ektverortung, on.
- the memory device SPE in which the landmark reference data LM-RD and the reference object data ROD are stored, is not contained in the vehicle control unit FZSE, but is located outside the vehicle control unit FZSE and z. B. is implemented as cloud storage (dashed representation of the storage device SPE in FIG. 1).
- the vehicle control unit FZSE according to option "B” thus only contains the control device STE, which in this state of the external storage device SPE j edoch assigned for data access . This assignment is shown in FIG. 1 by the dashed double arrow from the storage device SPE to the control device STE with the now uploaded computer-implemented tool CIW for carrying out the obstacle location in the vehicle control unit FZSE.
- the vehicle FZ can be any vehicle on land, water or air, provided that the vehicle area FZB is characterized by a route for the vehicle in question, by Landmarks and, if necessary, by additional reference objects and by obstacle objects that can pose a risk to the vehicle moving on the route.
- a rail vehicle e.g. B. Train with railcars, called .
- the vehicle area is a rail vehicle area that is defined by a track section for the rail vehicle, also by landmarks of the type defined at the outset, by reference objects, such as z. B. Catenary masts, as well as again by ects Obstj is marked.
- Characteristics resulting automatically from the algorithm e.g. B. S I FT features.
- Locating the obstacle in the vehicle FZ using the vehicle sensor system FZS and the vehicle control unit FZSE begins with the at least one sensor SS of the vehicle sensor system FZS being detected in the vehicle FZ by the route area FSB with the at least one landmark LM and the at least one obstacle object HO primary sensor measurement data SMDi and secondary sensor measurement data SMD2 recorded.
- the primary sensor measurement data SMDi belong to at least one sensor measurement object SMO L M, through which one of the at least one landmark LM could be represented, while the secondary sensor measurement data SMD2 belong to at least one sensor measurement object SMOHO, through which one in each case of the at least one obstacle object HO could be represented.
- This detection is sensor-related and independent of whether the vehicle FZ is moving or stationary, dependent on the distance.
- This fact is indicated in FIGURE 1 by the fact that the acquisition of the respective sensor measurement data is shown on the one hand with solid connecting lines between the sensor measurement objects and the vehicle sensor system and on the other hand with dashed connection lines between the sensor measurement objects and the vehicle sensor system.
- the control program commands of the Program module PGM for obstacle object j ektverortung leading processor PZ of the computer-implemented tool CIW if this has been uploaded to the control device STE, so implemented in the control device STE, on which the at least one sensor SS of the vehicle sensor system FZS Detected primary sensor measurement data SMDi and secondary sensor measurement data SMD2 by using these data z. B. read in or uploaded by the vehicle sensor system FZS.
- map material KM With regard to the map material KM, it is also possible for the map material KM, which contains the landmark reference data LM-RD, to be evaluated and verified in advance and also during operation, d. H . you can declare landmarks, areas and landmark patterns, which often lead to false associations, as unsafe information or take artificial measures in the vicinity of the route area FSB, such as e.g. B. the assembly of additional markers . This is done by a human observer or alternatively by an automated evaluation of the observation frequency in connection with a parameterizable threshold.
- the obstacle object j ektortung can perform.
- the aforementioned data SMDi, SMD2, LM-RD are entered into the processor PZ.
- the primary sensor measurement data SMDi and secondary sensor measurement data SMD2 are associated with the landmark reference data LM-RD representing the at least one landmark LM for landmark assignment DAZ.
- the result of this association is the following : a .
- the data association DAZ between the primary sensor measurement data SMDi and the landmark reference data LM-RD occurs and, as a result, a landmark assignment for the at least one sensor measurement object SMO LM is possible through associated primary sensor measurement data SMDi , az , whereby the respective landmark LM is represented by the associated primary sensor measurement data SMDi, az .
- the processor PZ and the program module PGM are preferably designed for locating the obstacle object in such a way that the data association DAZ for assigning landmarks is carried out using probabilistic factor graph methods or assignment/optimization methods.
- the probabilistic factor graph methods are preferably based on the implementation of software libraries such.
- B. “Georgia Tech Smoothing and Mapping ⁇ GTSAM>” or “ ⁇ g2o> as a General Framework for Graph Optimi zation”
- mapping/optimization methods refer to methods such as B. "Global Nearest Neighbor ⁇ GNN>", Munkres algorithms, "Joint Probalistic Data Association ⁇ JPDA>” or "Multi-Hypothesis-Tracking ⁇ MHT>”.
- the assignment does not only take place in one step, but is filtered over several time steps.
- the processor PZ and the program module PGM are designed accordingly for this purpose. This avoids misassociations leading to large errors.
- the assignment/optimization methods are alternative methods for data association, but poor performance is to be expected.
- some of the sensor measurement objects are assigned landmarks, some are assigned dynamic sensor measurement objects from previous journals and some cannot be assigned. These unassigned sensor measurement objects are relevant for the following steps.
- Sensor acquisition-specific localization distances VOD are determined between the at least one unassignable sensor measurement object SMOHO and the at least one landmark LM represented by the respective associated primary sensor measurement data SMDi, az on the basis of the associated primary sensor measurement data SMDi, az and the secondary sensor measurement data SMD2, by relating sensor detection-specific information IF ses contained in the associated primary sensor measurement data SMDi, az and in the secondary sensor measurement data SMD2 to one another in relation to the localization distances VOD to be determined.
- the sensor-acquisition-specific information IF ses is related to one another in relation to the location distances VOD to be determined in order to determine the location distances VOD sensor-acquisition-specifically, as a function from the type of measurement data acquisition of the at least one sensor SS of the vehicle sensor system FZS.
- the radar sensor based on the "Radio Detection and Ranging ⁇ Radar>” principle or the Lidar sensor based on the "Light Detection and Ranging ⁇ Lidar>” principle is used, Cartesian coordinate information KI F-kar, spherical coordinate information KI F sp h or cylindrical coordinate information KI F zyi in the associated primary sensor measurement data SMDi, az and in the secondary sensor measurement data SMD2 in relation to the localization distances to be determined VOD by angle difference f ferenzdetermination or determination of a Euclidean distance related to the To determine localization distances VOD between the at least one unassignable sensor measurement object SMOHO and the at least one landmark LM represented by the respective associated primary sensor measurement data SMDi, az
- the at least one sensor SS is based on the imaging principle SS B GP and is therefore z. B. the camera sensor is used, pixel information PI F in the associated primary sensor measurement data SMDi, az and in the secondary sensor measurement data SMD2 are related to one another in relation to the localization distances VOD to be determined by determining the pixel spacing in order to determine the localization distances VOD between the at least one to determine unassignable Sensormessobj ect SMOHO and the at least one by the respective associated primary sensor measurement data SMDi, az represented landmark LM.
- the at least one obstacle object HO is located on the route FS in the route area FSB, where it represents an obstacle for the vehicle FZ, VO and localization information VOI F is preferably generated or generated if either on the basis of the determined sen- sensor acquisition-specific localization distances VOD or on the basis of the determined sensor acquisition-specific localization distances VOD and using the landmark reference data LM-RD and/or the reference object data ROD, the at least one unassignable sensor measurement object SMOHO of the route FS can be allocated. Otherwise, if the at least one unassignable sensor measurement object SMOHO cannot be assigned to the route FS but can be assigned to the route area FSB, it does not represent an obstacle for the vehicle FZ. However, it is still located, specifically outside the route of the vehicle FZ in the route area FSB. Both cases are shown in FIG. 1 as an example.
- the generated localization information VOI F can z. B. be used to control measures such as e.g. B. automatically initiate a reduction in speed and, if necessary, braking of the vehicle, or z. B. to inform the vehicle driver with a warning about the localized obstacle object.
- control measures such as e.g. B. automatically initiate a reduction in speed and, if necessary, braking of the vehicle, or z. B. to inform the vehicle driver with a warning about the localized obstacle object.
- the processor PZ accesses it again, as in the case of the landmark reference data LM-RD, in that this data z. B. read from the storage device SPE according to option "A” or uploaded from the storage device SPE according to option "B".
- the processor PZ and the program module PGM are also designed for locating the obstacle object in such a way that locating VO the at least one obstacle object HO and, if necessary, generating or creating the location information with the sensor measurement data acquisition, the data association and the location distance determination at least in relation on the sensor measurement data acquisition and data association is carried out dynamically at different localization times in the sense of an object tracking. By tracking in particular the localization distance over time, their robustness and meaningfulness are increased. Alternatively, the sensor measurement objects can also be tracked and the localization distance can then be determined.
- the obstacle object HO that cannot be assigned and is located on the route FS in FIG. 1 is detected to the right of the two left landmarks HO and to the left of the right landmark HO.
- the known position of the landmarks LM relative to the route FS can be used to determine that the unassigned obstacle object HO represents an obstacle.
- the terms “to the left of” and “to the right of” can be resolved by angle or pixel differences. If said obstacle object HO were to the right of the right-hand landmark LM, it would clearly not represent an obstacle, as is the case with the other obstacle object HO shown in FIG. 1, which also cannot be assigned.
- FIG. 2 shows a modified vehicle scenario for locating an obstacle in a route area FSB identified by landmarks LM with an obstacle object HO extending vertically to a route FS of the route area FSB.
- the obstacle HO extending vertically to the route FS is a maintenance vehicle on the route FS and it must be determined whether the route FS can be driven on vertically.
- a known landmark LM e.g. B. in the form of a bridge, which is known to be underpassable, was recognized and determined by localization distances, which are z. B.
- FIG. 3 shows a modified vehicle scenario for locating obstacles in a route area FSB marked by landmarks LM with a route FS in which the landmarks LM are arranged in a grouping or formation GFLMi, GFLM2. Because the landmarks LM are not considered individually, but rather the grouping and formation of the landmarks GFLMi, GFLM2, the robustness of the obstacle object localization is increased.
- the landmark reference data LM-RD contain data that represent the grouping or formation of the landmarks GFLMi, GFLM2 for a group-based or formation-based landmark assignment and supplement the stored map material KM.
- the grouping and formation of the landmarks GFLMi, GFLM2 also simplifies the data association between the primary sensor measurement data SMDi or the secondary sensor measurement data SMD2 and the landmark reference data LM-RD.
- the landmark groupings or landmark formations are statically selected and defined in advance in the map material, so that only the data association from sensor measurement data to the landmark groupings or landmark formations has to take place for the obstacle object localization. In this case, a check is then carried out on detected sensor measurement objects as to whether they meet the properties of one of the known landmark groupings or landmark formations, which greatly simplifies the landmark assignment.
- FIG. 4 shows another vehicle scenario for locating obstacles based on the scenario according to FIG. 3, with landmarks in the route area being detected at different times t1, t2.
- the vehicle scenario for locating obstacles can be expanded in such a way that the landmark groupings or landmark formations GFLMi, GFLM2 are learned in the course of a dynamic ect localization of obstacles at the various localization times tl, t2, ie during operation, by the localization time tl
- Another landmark grouping or landmark formation GFLM12 created at the localization time t2 is added to the landmark groupings or landmark formations GFLMi, GFLM2 known from FIG.
- the landmark reference data LM-RD contain further data that represent the further landmark grouping or landmark formation GFLM12 for a group-based or formation-based landmark assignment and additionally supplement the stored map material KM.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Navigation (AREA)
- Traffic Control Systems (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102020210059.4A DE102020210059A1 (de) | 2020-08-07 | 2020-08-07 | Verfahren, Computer-implementiertes Werkzeug, Fahrzeug-Steuerungseinheit und Fahrzeug zum Verorten von Hindernisobjekten in landmarkengekennzeichneten Fahrstreckenbereichen von Fahrzeugen |
| PCT/EP2021/069758 WO2022028848A1 (de) | 2020-08-07 | 2021-07-15 | Verfahren, computer-implementiertes werkzeug, fahrzeug-steuerungseinheit und fahrzeug zum verorten von hindernisobjekten in landmarkengekennzeichneten fahrstreckenbereichen von fahrzeugen |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4154166A1 true EP4154166A1 (de) | 2023-03-29 |
Family
ID=77071514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP21746412.2A Pending EP4154166A1 (de) | 2020-08-07 | 2021-07-15 | Verfahren, computer-implementiertes werkzeug, fahrzeug-steuerungseinheit und fahrzeug zum verorten von hindernisobjekten in landmarkengekennzeichneten fahrstreckenbereichen von fahrzeugen |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US12330632B2 (de) |
| EP (1) | EP4154166A1 (de) |
| DE (1) | DE102020210059A1 (de) |
| WO (1) | WO2022028848A1 (de) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102020210059A1 (de) * | 2020-08-07 | 2022-02-10 | Siemens Mobility GmbH | Verfahren, Computer-implementiertes Werkzeug, Fahrzeug-Steuerungseinheit und Fahrzeug zum Verorten von Hindernisobjekten in landmarkengekennzeichneten Fahrstreckenbereichen von Fahrzeugen |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4428208B2 (ja) * | 2004-11-16 | 2010-03-10 | 株式会社デンソー | 車両用物体認識装置 |
| DE102008007347A1 (de) | 2008-02-04 | 2009-08-06 | Robert Bosch Gmbh | Vorrichtung und Verfahren zum Bestimmen der Position eines anderen Verkehrsteilnehmers |
| US8798841B1 (en) * | 2013-03-14 | 2014-08-05 | GM Global Technology Operations LLC | System and method for improving sensor visibility of vehicle in autonomous driving mode |
| DE102016205870A1 (de) * | 2016-04-08 | 2017-10-12 | Robert Bosch Gmbh | Verfahren zur Bestimmung einer Pose eines wenigstens teilautomatisiert fahrenden Fahrzeugs in einer Umgebung mittels Landmarken |
| JP6649191B2 (ja) * | 2016-06-29 | 2020-02-19 | クラリオン株式会社 | 車載処理装置 |
| DE102016217330A1 (de) * | 2016-09-12 | 2018-03-15 | Volkswagen Aktiengesellschaft | Verfahren zum Betreiben eines Fahrzeugs und Steuergerät zur Durchführung des Verfahrens |
| DE102016015405A1 (de) | 2016-12-22 | 2017-07-06 | Daimler Ag | Umfassende Umgebungserfassung für einen Kraftwagen mittels Radar |
| EP3454079B1 (de) * | 2017-09-12 | 2023-11-01 | Aptiv Technologies Limited | Verfahren zur bestimmung der angemessenheit eines radarziels als ein positionsorientierungspunkt |
| KR102727540B1 (ko) * | 2019-04-26 | 2024-11-11 | 현대모비스 주식회사 | 주차 지원 장치 및 방법 |
| DE102020210059A1 (de) * | 2020-08-07 | 2022-02-10 | Siemens Mobility GmbH | Verfahren, Computer-implementiertes Werkzeug, Fahrzeug-Steuerungseinheit und Fahrzeug zum Verorten von Hindernisobjekten in landmarkengekennzeichneten Fahrstreckenbereichen von Fahrzeugen |
-
2020
- 2020-08-07 DE DE102020210059.4A patent/DE102020210059A1/de not_active Withdrawn
-
2021
- 2021-07-15 WO PCT/EP2021/069758 patent/WO2022028848A1/de not_active Ceased
- 2021-07-15 EP EP21746412.2A patent/EP4154166A1/de active Pending
- 2021-07-15 US US18/017,972 patent/US12330632B2/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| US12330632B2 (en) | 2025-06-17 |
| US20230264688A1 (en) | 2023-08-24 |
| DE102020210059A1 (de) | 2022-02-10 |
| WO2022028848A1 (de) | 2022-02-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE102010006828B4 (de) | Verfahren zur automatischen Erstellung eines Modells der Umgebung eines Fahrzeugs sowie Fahrerassistenzsystem und Fahrzeug | |
| DE102014223363B4 (de) | Verfahren und Vorrichtung zur Lokalisation eines Kraftfahrzeugs in einer ortsfesten Referenzkarte | |
| DE102009006113B4 (de) | Vorrichtung und Verfahren zur Sensorfusion mit dynamischen Objekten | |
| DE102019215403A1 (de) | Vorrichtung zur wegerzeugung an einer kreuzung und verfahren und vorrichtung zum steuern eines fahrzeugs an einer kreuzung | |
| DE102004047130B4 (de) | Umgebungsmodellierung für autonome mobile Systeme | |
| DE112020005275T5 (de) | Systeme und verfahren zur selektiven verzögerung eines fahrzeugs | |
| DE102017206433B4 (de) | Automatikfahrtsteuerungssystem eines mobilen Objektes | |
| DE102008041679A1 (de) | Vorrichtung und Verfahren zur erinnerungsbasierten Umfelderkennung | |
| EP3627386A1 (de) | Verfahren und vorrichtung zum bereitstellen eines umfeldabbildes eines umfeldes einer mobilen einrichtung und kraftfahrzeug mit einer solchen vorrichtung | |
| DE102011111440A1 (de) | Verfahren zur Umgebungsrepräsentation | |
| DE102015205133A1 (de) | Verfahren und Vorrichtung zum Ermitteln einer Bewegungsplanung für ein zumindest teilweise automatisiertes Fahren eines Fahrzeugs | |
| EP3063732B1 (de) | Situationsanalyse für ein fahrerassistenzsystem | |
| DE102021119871B4 (de) | Verfahren und Prozessorschaltung zum Betreiben einer automatisierten Fahrfunktion mit Objektklassifikator in einem Kraftfahrzeug, sowie Kraftfahrzeug | |
| DE102017201664A1 (de) | Verfahren zur Lokalisierung eines höher automatisierten Fahrzeugs in einer digitalen Karte | |
| DE102019213403A1 (de) | Verfahren zur sensorbasierten Lokalisation eines Egofahrzeuges, Egofahrzeug und ein Computerprogramm | |
| DE102016003261A1 (de) | Verfahren zur Selbstlokalisierung eines Fahrzeugs in einer Fahrzeugumgebung | |
| DE112020003551T5 (de) | Straßenkurvenerzeugung in Bildern der realen Welt als ein Verfahren zur Datenaugmentierung | |
| DE102020106304A1 (de) | Verfahren zum Steuern eines Fahrzeuges auf einem Betriebshof, Fahrt- Steuereinheit und Fahrzeug | |
| DE102020127315B4 (de) | System und Verfahren zum Annotieren von Automobilradardaten | |
| DE102017105721A1 (de) | Verfahren zur Identifikation wenigstens eines Parkplatzes für ein Fahrzeug und Parkplatzassistenzsystem eines Fahrzeugs | |
| WO2022028848A1 (de) | Verfahren, computer-implementiertes werkzeug, fahrzeug-steuerungseinheit und fahrzeug zum verorten von hindernisobjekten in landmarkengekennzeichneten fahrstreckenbereichen von fahrzeugen | |
| DE102018115895B4 (de) | Verfahren und System zur Erkennung von Hindernissen | |
| DE102021207609A1 (de) | Verfahren zur Kennzeichnung eines Bewegungsweges | |
| DE102021119319B4 (de) | Vorrichtung und Verfahren zur Erkennung und/oder Repräsentation einer Signalisierungseinheit | |
| DE102022000849A1 (de) | Verfahren zur Erzeugung einer Umgebungsrepräsentation für ein Fahrzeug |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20221221 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) | ||
| RAP3 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: SIEMENS MOBILITY GMBH |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
| 17Q | First examination report despatched |
Effective date: 20251126 |