CN-116639126-B - Vehicle autonomous lane change method, system, device and storage medium

CN116639126BCN 116639126 BCN116639126 BCN 116639126BCN-116639126-B

Abstract

The invention discloses a method, a system, a device and a storage medium for automatically changing lanes of a vehicle, and relates to the technical field of automatic driving. The method comprises the steps of obtaining road traffic information, constructing a lane change barrier topological relation of a vehicle according to the road traffic information, wherein the lane change barrier topological relation comprises current lane information and adjacent lane information, determining a lane change collision risk, a lane change front-back vehicle return and a vehicle surrounding traffic flow disturbance according to the lane change barrier topological relation, carrying out lane change decision according to the lane change collision risk, the lane change front-back vehicle return and the vehicle surrounding traffic flow disturbance, wherein the lane change decision comprises a target lane, carrying out transverse and longitudinal joint planning constraint according to the lane change barrier topological relation and vehicle parameters based on the lane change decision, and carrying out secondary dynamic planning according to the initial lane change track to obtain a smooth track in the lane change process. The invention can improve the channel changing efficiency and the safety and reasonably plan the channel changing track.

Inventors

FU XIANG
PEI CHAO
WANG WENJU
JIANG XUELIANG
WANG JIJIE

Assignees

武汉理工大学

Dates

Publication Date: 20260512
Application Date: 20230612

Claims (10)

1. An autonomous lane change method for a vehicle, comprising the steps of: Acquiring road traffic information; Constructing a lane change barrier topological relation of a vehicle according to the road traffic information, wherein the lane change barrier topological relation comprises current lane information and adjacent lane information; determining lane change collision risk, lane change front and rear vehicle benefits and surrounding traffic flow disturbance according to the lane change obstacle topological relation; Carrying out lane change decision according to the lane change collision risk, the return of the own vehicle before and after lane change and the disturbance of traffic flow around the own vehicle, wherein the lane change decision comprises a target lane; based on the lane change decision, performing horizontal-longitudinal joint planning constraint according to the lane change obstacle topological relation and the vehicle parameter to obtain an initial lane change track; And carrying out secondary dynamic programming according to the initial track changing track to obtain a smooth track in the track changing process.
2. The method for automatically changing lanes of a vehicle according to claim 1, wherein said constructing a lane changing obstacle topology of a vehicle according to said road traffic information comprises the steps of: Screening current lane information according to the road traffic information, and constructing a current lane topological relation according to the current lane information, wherein the current lane topological relation comprises first obstacle information, first pilot vehicle information and first following vehicle information before and after a current lane; Screening adjacent lane information according to the road traffic information, and constructing an adjacent lane vehicle topological relation according to the adjacent lane information, wherein the adjacent lane vehicle topological relation comprises lane changing gap information, second pilot vehicle information of an adjacent lane and second following vehicle information; and determining the lane change obstacle topological relation of the vehicle according to the current lane topological relation and the adjacent lane vehicle topological relation.
3. The method for autonomous lane changing of a vehicle according to claim 2, wherein the determining a lane changing collision risk, a lane changing front-rear vehicle return and a vehicle surrounding traffic flow disturbance according to the lane changing obstacle topology relation comprises the steps of: Constructing a vehicle lane change risk potential field in front of and behind a vehicle in a target lane change gap according to the lane change obstacle topological relation; carrying out space-time track prediction according to the channel changing risk potential field to obtain channel changing collision risk of each space-time track point; Determining a preset pilot vehicle speed of a target lane, a vehicle speed of a vehicle under a current lane, a vehicle acceleration of the vehicle before and after lane changing of the target lane, a pilot vehicle acceleration of the vehicle after lane changing of the target lane, a vehicle acceleration after lane changing, a target lane changing gap size and a residual space in front of the vehicle according to the lane changing obstacle topological relation, and calculating a surrounding traffic disturbance of the vehicle; And determining the return of the vehicle before and after the lane change according to the vehicle speed under the current lane and the predicted vehicle speed under the target lane.
4. A method of autonomous lane change in a vehicle according to claim 3, wherein said lane change decision based on said lane change collision risk, said lane change front-to-rear vehicle return and said vehicle surrounding traffic flow disturbance comprises the steps of: Comparing the channel changing collision risk of the space-time track points with channel changing prediction to determine whether to change channels; and comparing the return of the own vehicle before and after the lane change with the disturbance of traffic flow around the own vehicle to determine whether the lane change is performed.
5. The method of autonomous lane changing of a vehicle of claim 4, wherein the surrounding traffic flow disturbance of the vehicle is calculated by the following equation: ; Wherein, the 、、 And Are all the preset weight coefficients, and the weight coefficients are all the same, For a preset pilot vehicle speed for the target lane, For the speed of the own vehicle in the current lane, The acceleration of the following vehicle after lane change for the following vehicle on the target lane, For the acceleration of the following vehicle before the lane change of the following vehicle on the target lane, The navigation vehicle acceleration after lane change for the target lane navigation vehicle, For the self-vehicle acceleration after the lane change, For the target lane-change gap size, Is the space left in front of the bicycle.
6. The method for autonomous lane changing of a vehicle according to claim 5, wherein said obtaining an initial lane changing trajectory based on said lane changing decision by performing a joint horizontal-vertical planning constraint according to said lane changing obstacle topology and said vehicle parameters comprises the steps of: based on the longitudinal planning track of the previous time node, combining the speed of the vehicle and a wheel friction elliptic algorithm to carry out transverse planning to obtain the curvature constraint of the transverse track; Based on a transverse planning result, carrying out longitudinal planning according to the related track obstacle information in the lane change obstacle topological relation to obtain longitudinal track speed curvature constraint; and respectively limiting a transverse track and a longitudinal track according to the transverse track curvature constraint and the longitudinal track speed curvature constraint to obtain an initial track change track.
7. The method for automatically changing lanes of a vehicle according to claim 6, wherein the step of performing the secondary dynamic programming according to the initial lane changing track to obtain a smooth track in the lane changing process comprises the following steps: Based on an S-L coordinate system, dispersing a road space of an initial lane change track into an ordered lattice at intervals of a distance, and marking lattice nodes as level nodes; each level node is traversed circularly, and collision detection is carried out on the current node and the obstacle; When the collision detection result is collision, discarding the level node; When the collision detection result is that no collision occurs, calculating a loss value of each previous node of the current level node, wherein the loss value comprises a transverse offset, a transverse change rate and a transverse deviation of an end position of the previous node and a target reference line; taking the relay node with the minimum loss value as the father node of the current level node; Traversing until the last level node, searching a father node through a linked list until the initial node is found to obtain a smooth track.
8. An autonomous lane-changing system for a vehicle, comprising: The first module is used for acquiring road traffic information; the second module is used for constructing a lane changing barrier topological relation of the vehicle according to the road traffic information, wherein the lane changing barrier topological relation comprises current lane information and adjacent lane information; The third module is used for determining lane change collision risk, lane change front and back vehicle profits and vehicle surrounding traffic flow disturbance according to the lane change obstacle topological relation; A fourth module, configured to make a lane change decision according to the lane change collision risk, the return of the own vehicle before and after the lane change, and the disturbance of traffic flow around the own vehicle, where the lane change decision includes a target lane; a fifth module, configured to perform horizontal-vertical joint planning constraint according to the lane change obstacle topology relationship and the vehicle parameter based on the lane change decision to obtain an initial lane change track; And a sixth module, configured to perform secondary dynamic planning according to the initial track change track to obtain a smooth track in the track change process.
9. An autonomous lane-changing apparatus for a vehicle, comprising: At least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the vehicle autonomous lane change method of any one of claims 1 to 7.
10. A computer-readable storage medium in which a processor-executable program is stored, characterized in that the processor-executable program is for implementing the vehicle autonomous lane change method according to any one of claims 1 to 7 when executed by the processor.

Description

Vehicle autonomous lane change method, system, device and storage medium Technical Field The present invention relates to the field of automatic driving technologies, and in particular, to a method, a system, a device, and a storage medium for automatically changing lanes of a vehicle. Background Unmanned technology is regarded as one of important solutions for driving safety and improving traffic efficiency as an important research topic in the traffic field. With the continuous progress of the internet of things technology, unmanned technology is also rapidly developed. In driving behaviors, the following driving and the lane changing driving are the most common two driving behaviors, and compared with the following driving, the automatic lane changing driving can improve the traffic efficiency to a certain extent, solve the problem of local traffic jam, and bring about greater potential safety hazard. An unreasonable lane change decision would directly bring the unmanned vehicle to a dangerous driving scenario. Therefore, the unmanned lane change decision research is very important, and meanwhile, how to effectively execute the lane change decision and generate a proper lane change path, so that the lane change process of the vehicle can be quickly, comfortably and safely completed, and the lane change process is also important content of lane change behavior research. The traditional lane change decision process is used for idealizing lane change behavior, the influence of lane change behavior of a vehicle on local traffic flow is not considered, and a lane change model only considers simple traffic scenes, and complex interaction and games among multiple vehicles are ignored, so that lane change efficiency is limited, namely, the lane change process possibly causes interference on the local traffic flow, and traffic passing efficiency and riding experience are influenced. In addition, the horizontal and vertical planning decoupling between traditional unmanned lane change decision plans can not effectively utilize lane change decision information, so that the planned lane change track can not meet the actual requirements. Disclosure of Invention The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a method, a system, a device and a storage medium for automatically changing the track of the vehicle, which can improve the track changing efficiency and the safety and reasonably plan the track changing track. In one aspect, an embodiment of the present invention provides a method for automatically changing lanes of a vehicle, including the following steps: Acquiring road traffic information; Constructing a lane change barrier topological relation of a vehicle according to the road traffic information, wherein the lane change barrier topological relation comprises current lane information and adjacent lane information; determining lane change collision risk, lane change front and rear vehicle benefits and surrounding traffic flow disturbance according to the lane change obstacle topological relation; Carrying out lane change decision according to the lane change collision risk, the return of the own vehicle before and after lane change and the disturbance of traffic flow around the own vehicle, wherein the lane change decision comprises a target lane; based on the lane change decision, performing horizontal-longitudinal joint planning constraint according to the lane change obstacle topological relation and the vehicle parameter to obtain an initial lane change track; And carrying out secondary dynamic programming according to the initial track changing track to obtain a smooth track in the track changing process. According to some embodiments of the invention, the constructing the road changing obstacle topology relation of the vehicle according to the road traffic information comprises the following steps: Screening current lane information according to the road traffic information, and constructing a current lane topological relation according to the current lane information, wherein the current lane topological relation comprises first obstacle information, first pilot vehicle information and first following vehicle information before and after a current lane; Screening adjacent lane information according to the road traffic information, and constructing an adjacent lane vehicle topological relation according to the adjacent lane information, wherein the adjacent lane vehicle topological relation comprises lane changing gap information, second pilot vehicle information of an adjacent lane and second following vehicle information; and determining the lane change obstacle topological relation of the vehicle according to the current lane topological relation and the adjacent lane vehicle topological relation. According to some embodiments of the present invention, the determining the lane change collision risk, the lane chan