Search

CN-121977587-A - Driving path generation method, driving path generation system, electronic device and computer readable medium

CN121977587ACN 121977587 ACN121977587 ACN 121977587ACN-121977587-A

Abstract

The invention provides a driving path generation method, which belongs to the field of intelligent driving and comprises the steps of obtaining current environment data of a vehicle; constructing a feasible region boundary according to the current environment data of the vehicle; searching the maximum passable radius according to the feasible region boundary; according to the method, the driving path is formed according to the maximum passable radius, so that the planned path is more suitable for human driving habits in the aspects of curvature change, steering time and the like, riding comfort can be improved, discomfort of passengers due to abrupt change of the path is reduced, other road users can more easily predict vehicle driving intention, and traffic safety is improved.

Inventors

  • SUN QINGJIE

Assignees

  • 东风汽车集团股份有限公司

Dates

Publication Date
20260505
Application Date
20260205

Claims (10)

  1. 1. A driving path generation method, characterized by comprising: Acquiring current environment data of a vehicle; Constructing a feasible region boundary according to the current environment data of the vehicle; Searching the maximum passable radius according to the feasible region boundary; And forming a driving path according to the maximum passable radius.
  2. 2. The driving path generating method according to claim 1, wherein in the step of acquiring the current environment data of the vehicle, comprising: collecting multi-sensor data; Preprocessing the multi-sensor data and performing time and space alignment; Extracting road boundaries, obstacles and lane line features from the multi-sensor data; And fusing the road boundary, the obstacle and the lane line characteristics to acquire the current environment data of the vehicle.
  3. 3. The driving path generating method according to claim 2, characterized in that in the step of constructing a viable area boundary from the vehicle current environment data, comprising: And constructing a feasible region boundary by adopting a grid map or geometric boundary representation or a semantic map according to the current environment data of the vehicle.
  4. 4. A driving path generating method according to claim 3, wherein in the step of finding a maximum passable radius from the feasible region boundary, comprising: Setting a search sampling boundary, wherein the lower limit is a straight running state corresponding to 0 curvature, and the upper limit is a minimum turning radius corresponding to the maximum curvature of the vehicle; Selecting an initial radius from the sampling boundary, and calculating an RS curve based on the initial radius; judging whether the calculated RS curve is completely in a drivable area, if so, continuing searching in a larger radius interval by setting the adjusting amplitude delta R, and if not, searching in a smaller radius interval by setting the adjusting amplitude delta R; the maximum passable radius is determined by iterative lookup and is labeled r_mid.
  5. 5. The driving path generating method according to claim 4, wherein in the step of forming the driving path according to the maximum passable radius, comprising: Calculating an RS curve of radii r_l and r_r, wherein r_l < r_mid < r_r, and r_mid-r_l=Δr, r_r-r_mid=Δr; The RS curves of r_mid, r_l and r_r are spliced into a continuous and smooth driving path.
  6. 6. The driving path generating method according to claim 5, wherein in the step of stitching RS curves of r_mid, r_l, and r_r into a continuous and smooth driving path, comprising: Determining the proportion of the curve segment using R_l to the curve segment using R_r according to environmental factors; And gradually adjusting the curvature of the curve section according to the road condition.
  7. 7. The driving path generating method according to claim 1, further comprising performing a boundary collision check on the driving path.
  8. 8. A driving path generation system configured to implement the method of any one of claims 1-7, comprising: the data acquisition unit is used for acquiring current environment data of the vehicle; The construction unit is used for constructing a feasible region boundary according to the current environment data of the vehicle; the searching unit is used for searching the maximum passable radius according to the feasible region boundary; And the generating unit is used for forming a driving path according to the maximum passable radius.
  9. 9. An electronic device, comprising: One or more processors; A memory for storing one or more programs; when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1 to 7.
  10. 10. A computer readable medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method according to any of claims 1 to 7.

Description

Driving path generation method, driving path generation system, electronic device and computer readable medium Technical Field The present invention relates to the field of intelligent driving technologies, and in particular, to a driving path generating method, a driving path generating system, an electronic device, and a computer readable medium. Background The RS curve, which is totally called Reeds-Shepp curve, is a core algorithm in the path planning field, allows the vehicle to advance and retreat, and solves the problem of shortest path planning from the starting point to the end point when the vehicle moves in a plane. However, RS curves still have limitations in practical applications: The theoretical fixed turning radius is greatly different from the actual turning radius of different vehicle types and robots, so that the error rate of a planned path exceeds 10%; the curvature of the connecting point of the adjacent curve sections of the RS curve is discontinuous, so that the riding comfort is reduced by about 20% when the automatic driving vehicle runs; In a complex environment, the calculated amount of path planning of an RS curve is increased sharply, and a certain urban road test shows that in a high-density traffic flow and complex intersection scene, the planning time is prolonged to 8-10 seconds, so that the real-time requirement is difficult to meet. Disclosure of Invention The invention aims to solve at least one technical problem existing in the prior art, and provides a driving path generation method and a driving path generation system. In a first aspect, an embodiment of the present invention provides a driving path generating method, including: Acquiring current environment data of a vehicle; Constructing a feasible region boundary according to the current environment data of the vehicle; Searching the maximum passable radius according to the feasible region boundary; And forming a driving path according to the maximum passable radius. In some of these embodiments, in the step of acquiring current environmental data of the vehicle, the method includes: collecting multi-sensor data; Preprocessing the multi-sensor data and performing time and space alignment; Extracting road boundaries, obstacles and lane line features from the multi-sensor data; And fusing the road boundary, the obstacle and the lane line characteristics to acquire the current environment data of the vehicle. In some of these embodiments, in the step of constructing a viable area boundary from the vehicle current environment data, it includes: And constructing a feasible region boundary by adopting a grid map or geometric boundary representation or a semantic map according to the current environment data of the vehicle. In some of these embodiments, the step of finding a maximum passable radius from the viable area boundary includes: Setting a search sampling boundary, wherein the lower limit is a straight running state corresponding to 0 curvature, and the upper limit is a minimum turning radius corresponding to the maximum curvature of the vehicle; Selecting an initial radius from the sampling boundary, and calculating an RS curve based on the initial radius; judging whether the calculated RS curve is completely in a drivable area, if so, continuing searching in a larger radius interval by setting the adjusting amplitude delta R, and if not, searching in a smaller radius interval by setting the adjusting amplitude delta R; the maximum passable radius is determined by iterative lookup and is labeled r_mid. In some of these embodiments, in the step of forming a driving path according to the maximum passable radius, it comprises: Calculating an RS curve of radii r_l and r_r, wherein r_l < r_mid < r_r, and r_mid-r_l=Δr, r_r-r_mid=Δr; The RS curves of r_mid, r_l and r_r are spliced into a continuous and smooth driving path. In some of these embodiments, in the step of stitching the RS curves of r_mid, r_l and r_r into a continuous and smooth driving path, it comprises: Determining the proportion of the curve segment using R_l to the curve segment using R_r according to environmental factors; And gradually adjusting the curvature of the curve section according to the road condition. In some embodiments, the method further comprises performing boundary collision verification on the driving path. In a second aspect, an embodiment of the present invention provides a driving path generating system configured to implement the above method, including: the data acquisition unit is used for acquiring current environment data of the vehicle; The construction unit is used for constructing a feasible region boundary according to the current environment data of the vehicle; the searching unit is used for searching the maximum passable radius according to the feasible region boundary; And the generating unit is used for forming a driving path according to the maximum passable radius. In a third aspect, an embodiment of the present inv