Search

CN-114537447-B - Secure passing method, apparatus, electronic device and storage medium

CN114537447BCN 114537447 BCN114537447 BCN 114537447BCN-114537447-B

Abstract

The embodiment of the disclosure discloses a safe passing method, a device, electronic equipment and a storage medium, wherein the method comprises the steps of acquiring a target obstacle in a preset distance range according to a vehicle-mounted sensor; the method comprises the steps of obtaining a blind area polygon of a perception blind area based on the target obstacle, obtaining a relation lane associated with a current lane of a host vehicle according to a high-precision map, obtaining a collision intersection point of the relation lane and the blind area polygon, determining the collision intersection point closest to the current lane of the host vehicle as a target intersection point, and determining the expected speed of the host vehicle according to the target intersection point so as to reduce the collision risk between the host vehicle and the obstacle in the perception blind area. The method and the device improve the running safety of the automatic driving vehicle in the running environment with the blind area and reduce the production cost.

Inventors

  • ZHANG ZHICHEN
  • TIAN XIAOSHENG
  • ZHANG SHUAI

Assignees

  • 驭势科技（北京）有限公司

Dates

Publication Date
20260512
Application Date
20220331

Claims (10)

  1. 1. A method of secure traffic, the method comprising: acquiring a target obstacle in a preset distance range according to the vehicle-mounted sensor; acquiring a blind area polygon of a perception blind area based on the target obstacle; obtaining a relation lane associated with a current lane of the vehicle according to the high-precision map; acquiring collision intersection points of the relation lanes and the blind area polygons, and determining the collision intersection point closest to the current lane of the vehicle as a target intersection point; Determining the expected speed of the vehicle according to the target intersection point so as to reduce the collision risk between the vehicle and the obstacle in the perception blind area; The obtaining the collision intersection point of the relation lane and the blind area polygon comprises the following steps: Acquiring sampling points corresponding to the relation lanes in a high-precision map; traversing all the sampling points based on a second preset algorithm to obtain at least one group of adjacent first sampling points and second sampling points, wherein the first sampling points are positioned outside the blind area polygon, and the second sampling points are positioned inside the blind area polygon; Acquiring a sample point ray based on the first sampling point and the second sampling point; And determining an intersection point of the sample point ray and the blind area polygon based on a collision detection algorithm, and determining the intersection point as a collision intersection point of the relation lane and the blind area polygon.
  2. 2. The method according to claim 1, wherein acquiring the target obstacle within the preset distance range according to the on-vehicle sensor comprises: Acquiring all barriers within a preset distance range through the vehicle-mounted sensor, and performing type filtering on all the barriers to acquire a filtered first barrier; and determining a first obstacle with a height larger than the height of the vehicle in the first obstacles as a target obstacle.
  3. 3. The method according to claim 1 or 2, wherein the obtaining a blind zone polygon of a perception blind zone based on the target obstacle comprises: Acquiring a tangent point of the target obstacle and a tangent line passing through the tangent point from the vehicle-mounted sensor based on a first preset algorithm according to the position relation between the target obstacle and the vehicle-mounted sensor; extending the tangent line by a preset length along the direction away from the vehicle to obtain a vertex; and determining the blind area polygon of the perception blind area according to the tangent point and the vertex.
  4. 4. The method of claim 1, wherein said determining a desired speed of the host vehicle from the target intersection comprises: acquiring a collision position of the intersection of the current lane and the relation lane of the vehicle; Acquiring the maximum deceleration of the vehicle, the first distance from the first current position of the vehicle to the collision position and the second distance from the target intersection point to the collision position; A desired speed of the host vehicle is determined based on a relationship of related quantities under defined conditions, wherein the related quantities include a speed of an assumed obstacle, a length of the assumed obstacle, a maximum deceleration of the host vehicle, the first distance, and the second distance.
  5. 5. The method of claim 4, wherein the defining conditions are: When the vehicle uniformly decelerates from the first current position to the collision position at the maximum deceleration, the assumed obstacle uniformly reaches a first position from the target intersection point position at the speed of the assumed obstacle; The distance from the vehicle to the collision position at the maximum deceleration from the first current position is the first distance; The distance that the assumed obstacle reaches the first position from the target intersection point position at a uniform speed is the sum of the second distance and the length of the assumed obstacle.
  6. 6. The method of claim 5, wherein the correlation of the correlation under the defined condition is: obj_t = ego_max_dec_t = ego_dist_to_intersect = ego_expect_vel * min(obj_t, ego_max_dec_t) +0.5*ego_max_dec*min(obj_t,ego_max_dec_t)*min(obj_t,ego_max_dec_t) Wherein ego _dist_to_ intersect represents the first distance, obj_dist_to_ intersect represents the second distance, obj_length represents the length of the hypothetical obstacle, ego _expect_vel represents the desired speed of the host vehicle, obj_vel represents the speed of the hypothetical obstacle, ego _max_dec represents the maximum deceleration of the host vehicle, ego _max_dec_t represents the time it takes for the host vehicle to travel the first distance from the first current position, obj_t represents the time it takes for the hypothetical obstacle to reach the first position at a uniform speed from the target intersection position.
  7. 7. The method according to any one of claims 4-6, wherein said determining the desired speed of the host vehicle based on the relation of the correlation amounts under the defined conditions comprises: Obtaining a first maximum speed limit of the vehicle reaching the collision position based on the time taken by the speed of the assumed obstacle to reach the first position from the target intersection point position at a uniform speed; obtaining a second maximum speed limit for the host vehicle to reach the collision position based on the first distance and a time taken for the host vehicle to travel the first distance from the first current position; And comparing the speed limit value of the first maximum speed limit and the second maximum speed limit, and determining the maximum speed limit with the minimum speed limit value as the expected speed.
  8. 8. A secure access device, comprising: the first acquisition module is used for acquiring a target obstacle in a preset distance range according to the vehicle-mounted sensor; The second acquisition module is used for acquiring a blind area polygon of a perception blind area based on the target obstacle; The first determining module is used for obtaining a relation lane associated with the current lane of the vehicle according to the high-precision map; The second determining module is used for obtaining collision intersection points of the relation lanes and the blind area polygons and determining the collision intersection point closest to the current lane distance of the vehicle as a target intersection point; The third determining module is used for determining the expected speed of the vehicle according to the target intersection point so as to reduce the collision risk between the vehicle and the obstacle in the perception blind area; the second determining module includes a third acquiring unit configured to: Acquiring sampling points corresponding to the relation lanes in a high-precision map; traversing all the sampling points based on a second preset algorithm to obtain at least one group of adjacent first sampling points and second sampling points, wherein the first sampling points are positioned outside the blind area polygon, and the second sampling points are positioned inside the blind area polygon; Acquiring a sample point ray based on the first sampling point and the second sampling point; And determining an intersection point of the sample point ray and the blind area polygon based on a collision detection algorithm, and determining the intersection point as a collision intersection point of the relation lane and the blind area polygon.
  9. 9. An electronic device, the electronic device comprising: one or more processors; a storage means for storing one or more programs; The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-7.
  10. 10. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any of claims 1-7.

Description

Secure passing method, apparatus, electronic device and storage medium Technical Field The disclosure relates to the technical field of automatic driving, and in particular relates to a safe passing method, a safe passing device, electronic equipment and a storage medium. Background In autopilot, the vehicle obtains environmental information entirely from onboard sensors, such as lidar, millimeter wave radar, cameras, and the like. Through the sensors, the information is transmitted to the vehicle-mounted core controller through a series of sensing algorithms, so that the automatic driving vehicle can run a series of decision planning algorithms according to the information, and the autonomous movement of the vehicle in a complex environment can be controlled. The mounting positions of the sensors are generally located at the top of the head of the vehicle and are farther than the distance seen by a human, but if the vehicle or the wall is arranged on the side, a part of shielding (blind area) is generated for the visual angle of the sensors, so that the sensors cannot observe the intersection at the left front or the right front, a certain safety risk is brought to the automatic driving vehicle, and if an obstacle with a higher speed suddenly appears in the blind area at the moment, the automatic driving vehicle can avoid untimely, thereby causing dangerous accidents. The driving of the human driver is also performed, a certain perception blind area exists, but the human driver can flexibly make some actions according to the environment, so that the driving can be safer and more reliable. However, if the vehicle is not perceived, the vehicle is considered safe at this time, and the environment cannot be understood like a person. In the existing solutions of the perceived blind areas in the automatic driving vehicles, for example, hundred-degree automatic driving adopts a vehicle-road cooperation technology, namely, a vehicle-road system device is installed at each traffic intersection, and a camera is arranged on each traffic intersection, so that the environment in the intersection can be perceived in real time, and information can be sent to the automatic driving vehicle which is about to enter the intersection. The method has two disadvantages, namely, the first method is to add a vehicle-road cooperation device at each intersection, which consumes manpower and material resources and increases cost, and the second method is that dead zones are not completely generated in the intersections, vehicles possibly stop at the sides of the intersections, a road extends out in front of the vehicles, dead zones are also generated, collision risks are caused, and at the moment, the vehicle-road cooperation technology cannot solve the problem. Disclosure of Invention In order to solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a safe passing method, an apparatus, an electronic device, and a storage medium, which improve the driving safety of an automatically driven vehicle in a driving environment with a blind area, and reduce the production cost. According to a first aspect, an embodiment of the disclosure provides a safe passing method, which comprises the steps of obtaining a target obstacle in a preset distance range according to a vehicle-mounted sensor, obtaining a blind area polygon of a perception blind area based on the target obstacle, obtaining a relation lane associated with a current lane of a vehicle according to a high-precision map, obtaining a collision intersection point of the relation lane and the blind area polygon, determining the collision intersection point closest to the current lane of the vehicle as the target intersection point, and determining the expected speed of the vehicle according to the target intersection point so as to reduce the collision risk between the vehicle and the obstacle in the perception blind area. The embodiment of the disclosure also provides a safe passing device, which comprises a first acquisition module, a second acquisition module, a first determination module, a second determination module and a third determination module, wherein the first acquisition module is used for acquiring a target obstacle in a preset distance range according to a vehicle-mounted sensor, the second acquisition module is used for acquiring a blind area polygon of a perception blind area based on the target obstacle, the first determination module is used for acquiring a relation lane associated with a current lane of a host vehicle according to a high-precision map, the second determination module is used for acquiring a collision intersection point of the relation lane and the blind area polygon, determining the collision intersection point closest to the current lane of the host vehicle as the target intersection point, and the third determination module is used for determining the expected sp