Method and apparatus for controlling an autonomous vehicle

Abstract

The invention relates to a method for controlling an autonomous vehicle (1) in the area of ​​an intersection (SK) using data (D) detected by a perimeter detection sensor system (4), according to which, before reaching the intersection (SK), a number of future vehicle positions are identified as potential stopping points along the driving path ahead of the vehicle (1) until reaching the intersection (SK). For each potential stopping point, the field of view (S) of the perimeter detection sensor system (4) is predicted taking into account the height profile of the road surface in the area of ​​the intersection (SK) and the potential stopping point for which the field of view (S) of the perimeter detection sensor system (4) is maximum is selected as a stopping position (POS) of the vehicle (1). At the stopping position (POS), the vehicle surroundings are detected by the perimeter detection sensor system (4) and the vehicle (1) is controlled to automatically enter the intersection (SK) depending on the traffic situation calculated from the data (D) of the perimeter detection sensor system (4). Furthermore, the invention also relates to a device (5) for controlling an autonomous vehicle (1) in the area of ​​an intersection (SK), comprising a perimeter detection sensor system (4).

Description

[Technical Field]

[0001] The present invention relates to a method for controlling an autonomously driven vehicle according to the preamble of claim 1.

[0002] The invention relates to a device for controlling an autonomously driven vehicle according to the preamble of claim 9. [Background technology]

[0003] From DE 10 2019 105 739 A1, a method for partially automatic guidance of a motor vehicle when the field of view of the surrounding sensor system of the motor vehicle is obstructed is known, which comprises the following steps: generating and outputting ease-in control signals for controlling lateral and longitudinal guidance of the vehicle to guide the vehicle into the intersection slowly; receiving surrounding signals indicative of the vehicle's surroundings while slowly approaching an intersection; - determining, based on the surrounding signals, whether the vehicle may continue to proceed slowly further into the intersection or whether it needs to stop or back up; generating and outputting control signals based on the determination to control lateral and longitudinal guidance of the vehicle to partially automatically guide the vehicle to continue slowly entering the intersection or to stop or reverse in accordance with the determination; Contains: Summary of the Invention [Problem to be solved by the invention]

[0004] The invention is based on the problem of providing a novel method for controlling an autonomously driven vehicle and a novel device for controlling an autonomously driven vehicle. [Means for solving the problem]

[0005] The above problem is solved according to the invention by a method having the features of claim 1 and by a device having the features of claim 9.

[0006] Advantageous embodiments of the invention are the subject of the dependent claims.

[0007] Methods for controlling automated, in particular highly automated or autonomous, vehicles in the area of ​​intersections use data detected by a surrounding detection sensor system.

[0008] According to the present invention, before reaching an intersection, a number of future vehicle positions are identified as potential stopping points along the vehicle's forward travel path until reaching the intersection. For each potential stopping point, the field of view of the surroundings detection sensor system is predicted taking into account the height profile of the road surface in the area of ​​the intersection, and the potential stopping point that maximizes the field of view of the surroundings detection sensor system is selected as the stopping position for the vehicle. At the stopping position, the surroundings of the vehicle are detected by the surroundings detection sensor system, and the vehicle is controlled to automatically enter the intersection according to the traffic conditions calculated from the data of the surroundings detection sensor system.

[0009] The method can be used to significantly improve perimeter detection in areas of intersections that are difficult for a vehicle's perimeter detection sensor system to recognize, for example, due to the road surface height profile, thereby improving safety when a vehicle automatically drives into and through an intersection.

[0010] In a possible embodiment of the method, at a stop position, the vehicle's active chassis changes its orientation in order to maximize the field of view of the surroundings detection sensor system, thereby further improving surroundings detection in the area of ​​the intersection.

[0011] In a further possible embodiment of the method, the position of the vehicle is changed by automated longitudinal and / or lateral movement at the stop position so as to maximize the field of view of the surroundings detection sensor system, which also allows for a further improvement of surroundings detection in the area of ​​the intersection.

[0012] In a further possible embodiment of the method, the potential stopping points are identified based on map data of a digital road map, which makes the identification of potential stopping points simple and reliable.

[0013] In a further possible embodiment of the method, the stopping point candidates are identified based on model data of the surrounding model, which again makes the identification of stopping point candidates simple and reliable.

[0014] In a further possible embodiment of the invention, the surroundings model is generated based on data from the surroundings detection sensor system, which advantageously allows the surroundings model to be updated in order to take into account possible changes in the vehicle's surroundings.

[0015] In a further possible embodiment of the present invention, a further three-dimensional surroundings model of the vehicle is created based on data detected at a stopped position by the surroundings detection sensor system around the vehicle, which three-dimensional surroundings model displays the vehicle surroundings with particularly high reliability and accuracy due to the wide field of view of the surroundings detection sensor system at a stopped position.

[0016] In a further possible embodiment of the method, the orientation of the vehicle is determined based on the further three-dimensional surroundings model and map data of a digital road map in order to maximize the field of view of the surroundings detection sensor system, thus allowing a particularly reliable and accurate representation of the vehicle's surroundings to maximize the expansion of the field of view in a particularly reliable manner.

[0017] In a further possible embodiment of the method, the vehicle is controlled to slowly approach the intersection, thereby reducing the risk of the vehicle colliding with other traffic participants. In this case, the term "slowly approach" can be interpreted in accordance with the understanding of a person skilled in the art, taking into account Section 8 of the German Road Traffic Act, which states in paragraph 2, paragraph 3 that "if visibility at a road point is poor and therefore visibility is not possible, the vehicle may approach the intersection or junction slowly and cautiously until visibility improves."

[0018] An apparatus for controlling an automatically driven vehicle in an intersection area comprises a perimeter detection sensor system. According to the invention, the apparatus comprises a processing unit designed to identify, before reaching the intersection, a number of future vehicle positions as potential stopping points along the vehicle's forward travel path until reaching the intersection. The processing unit is also designed to predict, for each potential stopping point, the field of view of the perimeter detection sensor system taking into account the height profile of the road surface in the area of ​​the intersection, and to select as the stopping position of the vehicle the stopping point that maximizes the field of view of the perimeter detection sensor system. The apparatus further comprises a control unit designed to control the vehicle to automatically enter the intersection depending on the traffic situation calculated from data about the vehicle's surroundings detected by the perimeter detection sensor system at the stopping positions.

[0019] The device can be used to significantly improve perimeter detection in areas of intersections that are difficult for the vehicle's perimeter detection sensor system to see, for example, due to the road surface height profile, thereby improving safety when vehicles are driving autonomously through and entering intersections.

[0020] In a possible embodiment of the device, the device comprises an active chassis that is designed to redirect the vehicle in a stopped position to maximize the field of view of the surroundings detection sensor system, thereby further improving surroundings detection in intersection areas.

[0021] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. [Brief explanation of the drawings]

[0022] [Figure 1] FIG. 2 is a schematic top view of an intersection. [Figure 2] FIG. 1 is a schematic perspective view of a plurality of roads. [Figure 3] 1A-1C are schematic front views of a vehicle in various states. [Figure 4] 1 is a schematic block diagram of an apparatus for controlling an autonomous vehicle; DETAILED DESCRIPTION OF THE INVENTION

[0023] In all the drawings, the same reference numerals are used to designate corresponding parts.

[0024] In FIG. 1, a top view of an intersection SK is shown where a vehicle 1 and two other vehicles 2, 3 are present.

[0025] The vehicle 1 is designed for automated, particularly highly automated or autonomous, driving. To perform automated driving, it is necessary to detect the vehicle's surroundings. For this detection, the vehicle 1 is equipped with a surroundings detection sensor system 4.

[0026] To ensure safe passage through intersection SK while avoiding collisions and complying with legal provisions, such as the right of way, all other traffic participants, in this case the other vehicles 2 and 3, must be detected before vehicle 1 passes through intersection SK (before it).

[0027] Figure 2 shows perspective views of several roads FB1-FB4, which are characterized at least in part by a hilly elevation profile of the road surface. Figure 3 shows front views of a vehicle 1 in various conditions on road FB1, which has an uneven surface.

[0028] The height profile of the roads FB1-FB4 and / or uneven road surfaces, e.g. potholes L and / or bumps E, may limit the field of view S of the sensors of the surroundings detection sensor system 4, which have a limited vertical opening angle. These limitations may result in insufficient detection of the road surface and insufficient inclination for the corresponding sensors to detect the surroundings of the vehicle and / or traffic participants moving along the roads FB1-FB4.

[0029] To avoid such limitations of the field of view S, a device 5 and a method for controlling the autonomous vehicle 1 are provided.

[0030] FIG. 4 shows a block diagram of a possible embodiment of such a device 5 .

[0031] The device 5 comprises a surroundings detection sensor system 4, which may include, for example, a camera sensor, a radar sensor and / or a lidar sensor, and / or other sensors for detecting the surroundings of the vehicle. The device 5 further comprises a processing unit 6 having a number of processing modules 6.1 to 6.4, a control unit 7, a digital road map 8, an active chassis 9 of the vehicle 1, a fusion module 10, and a control module 11.

[0032] The data D sensed by the multiple sensors of the surrounding detection sensor system 4 are fused by a fusion module 10 to form fused data FD, which are used to detect static and dynamic objects around the vehicle.

[0033] The three-dimensional surroundings model UM1 is generated by a first processing module 6.1 of the processing unit 6 from the data D detected by the surroundings detection sensor system 4 and the map data KD of the digital road map 8 using prediction of the road surface profile and road surface detection, e.g. potholes L and / or bumps E.

[0034] Next, the second processing module 6.2 of the processing unit 6 uses the surroundings model UM1 and the map data KD to calculate the stopping position POS of the vehicle 1 from among a plurality of stopping point candidates indicating future vehicle positions on the driving route ahead of the vehicle 1. At this time, for each stopping point candidate, the field of view S of the surroundings detection sensor system 4 is predicted taking into account the height profile of the road surface in the area of ​​the intersection SK, and the stopping point candidate that maximizes the field of view S of the surroundings detection sensor system 4 is selected as the stopping position POS of the vehicle 1.

[0035] This stopping position POS is transmitted to a control unit 7 for controlling the automated longitudinal and / or lateral movement of the vehicle 1, whereby the vehicle 1 is guided to the stopping position POS and stopped at that position.

[0036] At the stopping position POS, a possible field of view S of the surroundings detection sensor system 4 is calculated by a further processing module 6.3 depending on the map data KD, in particular on the road geometry of the intersection SK contained therein, and on the basis of the fusion data FD. In this case, a further three-dimensional surroundings model UM2 is calculated from the data D and / or the fusion data FD, on the basis of which the processing module 6.4 determines whether a change in the orientation of the vehicle 1 by the active chassis 9 of the vehicle 1 at the stopping position POS allows the field of view S of the surroundings detection sensor system 4 to be increased. If so, the orientation of the vehicle 1, i.e. the inclination of the vehicle 1 relative to the road surface, is changed by the active chassis 9 so that the field of view S is maximized.

[0037] Furthermore, at the stopping position POS, the processing module 6.4 checks, depending on the map data KD and the fusion data FD, whether a further change, in particular a small change, of the position of the vehicle 1 by automated longitudinal and / or lateral movements allows a further enlargement of the field of view S of the surrounding detection sensor system 4. If so, the vehicle 1 is moved from the stopping position POS to the new stopping position POS by the control unit 7 for controlling the automated longitudinal and / or lateral movements.

[0038] When the vehicle 1 reaches the stopping position POS with the maximum field of view S and the inclination of the vehicle 1 is appropriately adjusted by the active chassis 9, the surroundings detection sensor system 4 detects the surroundings of the vehicle at the stopping position POS. Depending on the traffic situation calculated from the data D and / or the fused data FD of the surroundings detection sensor system 4, the control module 11 automatically controls the vehicle 1 to enter the intersection SK. This can be done, for example, by entering slowly. [Explanation of symbols]

[0039] 1 vehicle 2 vehicles 3 vehicles 4 Surrounding detection sensor system 5 Equipment 6 Processing Unit 6.1~6.4 Processing Module 7. Control Unit 8 Digital Road Maps 9 Active Chassis 10 Fusion Module 11 Control Module D Data E-ridge FB1~FB4 Road FD fusion data KD map data L Pothole POS stop position S field of view SK Intersection UM1 ambient model UM2 ambient model [Prior art documents] [Patent documents]

[0040] [Patent Document 1] German Patent Application Publication No. 102019105739

Claims

1. A method for controlling an autonomous vehicle (1) in the area of ​​an intersection (SK) using data (D) detected by a surroundings detection sensor system (4), comprising: - before reaching said intersection (SK), a number of future vehicle positions are identified as potential stopping points along the travel path ahead of said vehicle (1) until reaching said intersection (SK), - for each of said potential stopping points, the field of view (S) of said surroundings detection sensor system (4) is predicted taking into account the height profile of the road surface in the area of ​​said intersection (SK); - the candidate stopping point where the field of view (S) of the surroundings detection sensor system (4) is maximized is selected as the stopping position (POS) of the vehicle (1); - at the stop position (POS), the surroundings of the vehicle are detected by the surroundings detection sensor system (4), - the vehicle (1) is controlled to automatically enter the intersection (SK) according to the traffic situation calculated from the data (D) of the surroundings detection sensor system (4); The stopping point candidates are identified based on model data of the surrounding model (UM1). A method characterized by:

2. 2. The method according to claim 1, characterized in that, at the stop position (POS), the orientation of the vehicle (1) is changed by an active chassis (9) of the vehicle (1) so that the field of view (S) of the surroundings detection sensor system (4) is maximized.

3. 3. The method according to claim 1 or 2, characterized in that at the stop position (POS) the position of the vehicle (1) is changed by automated longitudinal and / or lateral movement so that the field of view (S) of the surroundings detection sensor system (4) is maximized.

4. 3. The method according to claim 1 or 2, characterized in that the potential stopping points are identified on the basis of map data (KD) of a digital road map (8).

5. 2. The method according to claim 1, characterized in that the surroundings model (UM1) is generated on the basis of the data (D) of the surroundings detection sensor system (4).

6. 3. The method according to claim 1 or 2, characterized in that a further three-dimensional environment model (UM2) of the vehicle's surroundings is created based on the data (D) detected at the stopping position (POS) by the environment detection sensor system (4) of the vehicle's surroundings.

7. 7. The method according to claim 6, characterized in that the orientation of the vehicle (1) is determined based on the further three-dimensional surroundings model (UM2) and map data (KD) of a digital road map (8) so as to maximize the field of view (S) of the surroundings detection sensor system (4).

8. A device (5) for controlling a vehicle (1) operating automatically in the area of ​​an intersection (SK), comprising a surroundings detection sensor system (4), - before reaching said intersection (SK), identifying a number of future vehicle positions as potential stopping points along the travel path ahead of said vehicle (1) until reaching said intersection (SK), - for each of said potential stopping points, predicting the field of view (S) of said surroundings detection sensor system (4) taking into account the height profile of the road surface in the area of ​​said intersection (SK), - selecting the candidate stopping point where the field of view (S) of the surroundings detection sensor system (4) is maximized as the stopping position (POS) of the vehicle (1); a processing unit (6) designed to - Controlling the vehicle (1) so that it automatically enters the intersection (SK) according to the traffic situation calculated from the data (D) around the vehicle detected by the surroundings detection sensor system (4) at the stopping position (POS), a control unit (7) designed to It is equipped with The stopping point candidates are identified based on model data of the surrounding model (UM1).

10. An apparatus (5) according to claim 1, characterized in that:

9. 9. The device (5) according to claim 8, characterized in that it comprises an active chassis (9) designed to change the orientation of the vehicle (1) in the stopped position (POS) so that the field of view (S) of the surrounding detection sensor system (4) is maximized.

10. A method for controlling an autonomous vehicle (1) in the area of ​​an intersection (SK) using data (D) detected by a surrounding detection sensor system (4), comprising: - before reaching said intersection (SK), a number of future vehicle positions are identified as potential stopping points along the travel path ahead of said vehicle (1) until reaching said intersection (SK), - for each of said potential stopping points, the field of view (S) of said surroundings detection sensor system (4) is predicted taking into account the height profile of the road surface in the area of ​​said intersection (SK); - the candidate stopping point where the field of view (S) of the surroundings detection sensor system (4) is maximized is selected as the stopping position (POS) of the vehicle (1); - at the stop position (POS), the surroundings of the vehicle are detected by the surroundings detection sensor system (4), - the vehicle (1) is controlled to automatically enter the intersection (SK) according to the detection status of other vehicles (2, 3) calculated from the data (D) detected by the surroundings detection sensor system (4) at the stopping position (POS); At the stop position (POS), the inclination of the vehicle (1) relative to the road surface is changed by an active chassis (9) of the vehicle (1) so that the field of view (S) of the surroundings detection sensor system (4) is maximized. A method characterized by:

Images