CN-121973770-A - Collision risk detection method, device, equipment, medium and program product

Abstract

The present disclosure relates to a collision risk detection method, apparatus, device, medium, and program product. The intelligent driving method comprises the steps of obtaining obstacle information, a vehicle planning route and a target area, responding to the overlapping area of the obstacle information and the vehicle planning route, conducting obstacle intrusion detection on the overlapping area based on the target area, and responding to the fact that obstacle intrusion exists in the overlapping area, and starting an obstacle avoidance strategy of a vehicle. According to the method and the device for preventing the obstacle avoidance function of the vehicle, whether the obstacle invasion exists in the vehicle planning route is confirmed, and the obstacle avoidance strategy of the vehicle is started when the obstacle invasion exists in the vehicle planning route in the real state, so that the false starting of the obstacle avoidance function of the vehicle caused by the obstacle perception error can be avoided. By the method provided by the disclosure, the interference of the false start of the obstacle avoidance function of the vehicle to the user in the driving process can be effectively reduced, and the driving experience of the user is improved.

Inventors

• LIU CHANGJIANG
• WANG YUNHONG

Assignees

• 小米汽车科技有限公司

Dates

Publication Date

20260505

Application Date

20260316

Claims (16)

1. 1. A collision risk detection method, the method comprising: The method comprises the steps of obtaining obstacle information, a vehicle planning route and a target area, wherein the obstacle information is used for representing range information of obstacles on the vehicle planning route, and the target area is used for representing a drivable area of a road where the vehicle is located; In response to the obstacle information and the planned route of the vehicle having an overlapping region, performing obstacle intrusion detection on the overlapping region based on the target region; And in response to determining that the overlapping area has obstacle invasion, starting an obstacle avoidance strategy of the vehicle, wherein the obstacle avoidance strategy comprises an automatic emergency braking strategy and/or a bypass obstacle avoidance strategy.
2. 2. The method of claim 1, wherein performing obstacle intrusion detection on the overlapping region based on the target region comprises: Dividing the target area into a plurality of area grids, and determining a target area grid corresponding to the overlapped area from the area grids, wherein the target area grid is used for representing grids which are preliminarily judged to be passable by the vehicle; acquiring point cloud data of the target area grid, and determining a point cloud state of the target area grid based on the point cloud data; And screening and confirming the passable area grids which can be passed by the vehicle from the target area grids according to the point cloud state of the target area grids so as to perform obstacle intrusion detection on the overlapped area.
3. 3. The method of claim 2, wherein determining a target region mesh corresponding to the overlapping region from the region meshes comprises: matching a region grid corresponding to the overlapping region from the region grids; acquiring characteristic information of an area grid corresponding to the overlapped area, wherein the characteristic information comprises grid identification information of the area grid corresponding to the overlapped area; and screening the target area grids from the area grids corresponding to the overlapped area according to the grid identification information.
4. 4. The method of claim 3, wherein the characteristic information further comprises a road surface height of a target area grid, wherein determining a point cloud state of the target area grid based on the point cloud data comprises: Determining a first preset range and a second preset range of the target area grid according to the road surface height of the target area grid; Determining a first point cloud quantity of the target area grid in the first preset range and a second point cloud quantity of the target area grid in the second preset range according to the point cloud data; And determining the point cloud state of the target area grid according to the first point cloud quantity and the second point cloud quantity.
5. 5. The method of claim 4, wherein determining the first pre-set range and the second pre-cloud range of the target area grid based on the road surface height of the target area grid comprises: determining a first preset range according to the road surface height of the target area grid and a first preset height threshold; and determining the second preset range according to the road surface height of the target area grid and a second preset height threshold value, wherein the first preset height threshold value is smaller than the second preset height threshold value.
6. 6. The method of claim 4, wherein determining the point cloud status of the target area grid from the first and second point cloud quantities comprises: Determining the target area grid as the passable area grid in response to the first point cloud quantity being greater than a preset point cloud quantity threshold and the first point cloud quantity being consistent with the second point cloud quantity; and determining that the target area grid is a non-passable area grid in response to the first point cloud quantity being less than or equal to a preset point cloud quantity threshold and/or the first point cloud quantity being inconsistent with the second point cloud quantity.
7. 7. The method of claim 2, wherein screening the passable area grid from which the vehicle is passable to be confirmed from the target area grid according to the point cloud state of the target area grid to perform obstacle intrusion detection on the overlapping area comprises: screening the number of the passable area grids from the target area grids according to the point cloud state of the target area grids; Determining the number of the target area grids, and calculating the ratio between the number of the passable area grids and the number of the target area grids; And detecting the obstacle invasion in the overlapped area according to a ratio result, wherein if the ratio result is smaller than a preset ratio, the obstacle invasion in the overlapped area is determined, otherwise, the obstacle invasion in the overlapped area is determined not to exist.
8. 8. The method according to any one of claims 1-7, further comprising: Generating collision risk prompt information in response to determining that the overlap region has obstacle intrusion; And sending the collision risk prompt information to an auxiliary driving module.
9. 9. The method according to any one of claims 1-7, further comprising: Acquiring multi-mode data information, wherein the multi-mode data information is used for representing surrounding environment information of the vehicle driving direction and vehicle state information; Processing the multi-modal data based on at least one perception model to obtain an environmental perception result, wherein the environmental perception result comprises the obstacle information and/or the target area; And planning the vehicle driving path according to the environment perception result based on a path planning model so as to obtain the vehicle planning path.
10. 10. A collision risk detection apparatus, characterized in that the apparatus comprises: The system comprises an information acquisition unit, a vehicle planning unit and a target area, wherein the information acquisition unit is used for acquiring barrier information, a vehicle planning route and a target area, the barrier information is used for representing range information of barriers existing on the vehicle planning route, and the target area is used for representing a drivable area of a road where the vehicle is located; An intrusion detection unit configured to respond to existence of an overlapping area between the obstacle information and a planned route of the vehicle, and perform obstacle intrusion detection on the overlapping area based on the target area; The obstacle avoidance strategy starting unit is used for starting an obstacle avoidance strategy of the vehicle in response to the fact that the overlapping area is determined to have obstacle invasion, and the obstacle avoidance strategy comprises an automatic emergency braking strategy and/or a bypass obstacle avoidance strategy.
11. 11. The apparatus according to claim 10, wherein the intrusion detection unit is further configured to divide the target area into a plurality of area grids, determine a target area grid corresponding to the overlapping area from the area grids, characterize the mesh preliminarily determined to be passable by the vehicle, acquire point cloud data of the target area grid, determine a point cloud state of the target area grid based on the point cloud data, and screen the passable area grid determined to be passable by the vehicle from the target area grid according to the point cloud state of the target area grid to perform obstacle intrusion detection on the overlapping area.
12. 12. The apparatus of claim 11, wherein the intrusion detection unit is further configured to match an area mesh corresponding to the overlapping area from the area meshes, obtain feature information of the area mesh corresponding to the overlapping area, the feature information including mesh identification information of the area mesh corresponding to the overlapping area, and screen a target area mesh from the area mesh corresponding to the overlapping area according to the mesh identification information.
13. 13. The apparatus of claim 12, wherein the characteristic information further comprises a road surface height of a target area grid, wherein the intrusion detection unit is further configured to determine a first preset range and a second preset range of the target area grid according to the road surface height of the target area grid, determine a first number of point clouds of the target area grid within the first preset range and a second number of point clouds within the second preset range according to the point cloud data, and determine a point cloud state of the target area grid according to the first number of point clouds and the second number of point clouds.
14. 14. A vehicle, characterized by comprising: A processor; A memory for storing processor-executable instructions; wherein the processor is configured to implement the collision risk detection method of any one of claims 1 to 9.
15. 15. A non-transitory computer readable storage medium, which when executed by a processor of a terminal, enables the terminal to perform the collision risk detection method of any one of claims 1 to 9.
16. 16. A computer program product comprising a computer program which, when executed by a processor, implements the collision risk detection method according to any one of claims 1 to 9.

Description

Collision risk detection method, device, equipment, medium and program product Technical Field The present disclosure relates to the field of intelligent driving technology, and in particular, to a collision risk detection method, apparatus, electronic device, non-transitory computer readable storage medium, and computer program product. Background Active safety is used as a basic function of vehicle auxiliary driving, and is a guarantee of high-order auxiliary driving safety. Specifically, the active safety can intervene in the vehicle dynamics through an electronic control system to prevent the accident, and the core aim is to prevent the wheel slip, steering out of control or lane departure through the cooperation of the sensor monitoring and the executing mechanism. In recent years, with the great development of driving assistance technology, active safety has also been greatly expanded in a functional effective scene. Therefore, when the triggering capability of active safety is improved, the false start initiation rate is not increased, and the driving experience of a user is prevented from being influenced by false braking. Disclosure of Invention To overcome the problems in the related art, the present disclosure provides a collision risk detection method, apparatus, electronic device, non-transitory computer-readable storage medium, and computer program product. According to a first aspect of an embodiment of the present disclosure, a collision risk detection method is provided, and the method includes obtaining obstacle information, a vehicle planning route and a target area, wherein the obstacle information is used for representing range information of obstacles existing on the vehicle planning route, the target area is used for representing a drivable area of a road where the vehicle is located, performing obstacle intrusion detection on the overlapped area based on the target area in response to the obstacle information and the existence of the overlapping area of the vehicle planning route, and starting an obstacle avoidance strategy of the vehicle in response to the fact that the overlapped area is confirmed to exist obstacle intrusion, and the obstacle avoidance strategy includes an automatic emergency braking strategy and/or a detour obstacle avoidance strategy. The embodiment of the disclosure confirms whether the obstacle invasion exists in the vehicle planning route truly, and starts the obstacle avoidance strategy of the vehicle when confirming that the obstacle invasion exists in the vehicle planning route truly, so that false starting of the obstacle avoidance function of the vehicle caused by obstacle perception errors can be avoided. In some exemplary embodiments of the disclosure, performing obstacle intrusion detection on the overlapping area based on the target area includes dividing the target area into a plurality of area grids, determining a target area grid corresponding to the overlapping area from the area grids, wherein the target area grid is used for representing grids which are preliminarily judged to be passable by the vehicle, acquiring point cloud data of the target area grid, determining a point cloud state of the target area grid based on the point cloud data, and screening passable area grids which are confirmed to be passable by the vehicle from the target area grids according to the point cloud state of the target area grid so as to perform obstacle intrusion detection on the overlapping area. According to the method and the device for detecting the obstacle frame, the secondary verification is carried out on the invasion of the obstacle frame through the road surface target area, the accuracy of the obstacle entering judgment is further improved, and various false detection and inaccurate detection situations can be effectively filtered under the guarantee of the high accuracy. In some exemplary embodiments of the present disclosure, determining a target area grid corresponding to the overlapping area from the area grids includes matching an area grid corresponding to the overlapping area from the area grids, acquiring feature information of the area grid corresponding to the overlapping area, the feature information including grid identification information of the area grid corresponding to the overlapping area, and screening the target area grid from the area grids corresponding to the overlapping area according to the grid identification information. According to the embodiment of the disclosure, the target area grids are screened from the area grids corresponding to the overlapping area through the characteristic information to serve as a preliminary screening result, and on the basis, the point cloud data of the target area grids are acquired again to carry out secondary verification, so that the data quantity of the point cloud data required to be acquired can be reduced, the efficiency of the secondary verification can be improved on the basis of reducin