CN-121977557-A - Path planning algorithm and system for four-wheel independent steering vehicle

CN121977557ACN 121977557 ACN121977557 ACN 121977557ACN-121977557-A

Abstract

The application belongs to the technical field of vehicle path planning. A path planning algorithm for a four-wheel independent steering vehicle comprises the steps of obtaining a vehicle parameter set, obtaining environment information of the vehicle, obtaining the environment information set in the environment information, carrying out discretization processing on a map according to the environment information set and the vehicle parameter set, expanding obstacles according to the size of the vehicle, calculating the distance from each position in the map to an end point node from the end point node in the map to obtain Dijkstra distance values of each point on the discrete map, carrying out grid sampling on ICR of the vehicle according to the vehicle parameter set, forming a motion model of the vehicle meeting the limitation of the change rate of steering angle of the wheel to obtain ICR grid sampling results, and obtaining the path planning of the vehicle according to the vehicle parameter set, the environment information set, the Dijkstra distance values of each point on the discrete map and the ICR grid sampling results.

Inventors

HU RUI
LIU YUFEI
WANG JIANWEI

Assignees

赛诺威盛科技(北京)股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260113

Claims (10)

1. A path planning algorithm for a four-wheel independent steering vehicle, the method comprising: Acquiring basic parameters of a vehicle, and determining a vehicle parameter set under a coordinate system of the vehicle, wherein the vehicle parameter set comprises the size of the vehicle, the position coordinates of four wheels of the vehicle, the distances from the four wheels to the center of the vehicle, a time step of path planning and the maximum variation allowed by the steering angle of the wheels in one time step; Acquiring environment information of the vehicle, and obtaining an environment information set in the environment information, wherein the environment information set comprises the position of an obstacle, the boundary of a map, and a starting point node and an end point node of the vehicle; performing discretization processing on the map according to the environment information set and the vehicle parameter set, expanding the obstacle according to the size of the vehicle, and calculating the distance from each position in the map to the terminal point from the terminal point in the map to obtain Dijkstra distance values of each point on the discrete map; According to the vehicle parameter set, carrying out grid sampling on ICR of the vehicle, and forming a motion model of the vehicle meeting the change rate limit of the steering angle of the wheel to obtain an ICR grid sampling result, wherein the ICR grid sampling result comprises all sampling points and adjacent relations thereof; and obtaining the path planning of the vehicle according to the vehicle parameter set, the environment information set, dijkstra distance values of each point on the discrete map and the ICR grid sampling result.
2. The path planning algorithm for a four-wheel independently steered vehicle according to claim 1, wherein the step of acquiring basic parameters of the vehicle and determining a set of vehicle parameters in a coordinate system in which the vehicle is located comprises: acquiring the outline dimension of the vehicle and the position coordinates of the wheels, and acquiring the outline dimension of the vehicle and the position coordinates of the four wheels; calculating the distance from each wheel to the center of the vehicle by using a Euclidean distance formula according to the position coordinates of the four wheels to obtain the distance from the wheel to the center of the vehicle; And determining the time step of path planning and the maximum variation allowed by the steering angle of the wheels in one time step according to the preset motion performance requirement or constraint of the vehicle, and obtaining the maximum variation of the time step and the steering angle.
3. The path planning algorithm of four-wheel independent steering vehicle according to claim 1, wherein the step of discretizing the map according to the set of environmental information and the set of vehicle parameters, expanding the obstacle according to the size of the vehicle, and calculating the distance from each position in the map to the destination node from the destination node in the map to obtain Dijkstra distance values of each point on the discrete map comprises: converting the map into a discrete grid structure according to the map size range in the environment information set to obtain a discrete grid map; Expanding the obstacle according to the position of the obstacle in the environment information set and the size of the vehicle in the vehicle parameter set to obtain the expanded discrete grid map; And in the expanded discrete grid map, converting the end point nodes in the environment information set into grid indexes to form an end point grid, and executing a Dijkstra algorithm by taking the end point grid as a starting point to obtain a Dijkstra distance value D of each point on the discrete map.
4. The path planning algorithm for a four-wheel independent steering vehicle according to claim 1, wherein the step of performing grid sampling on the ICR of the vehicle according to the vehicle parameter set and forming a motion model of the vehicle meeting the wheel steering angle change rate limit to obtain an ICR grid sampling result includes: According to the distances from the four wheels in the vehicle parameter set to the center of the vehicle and the maximum variation allowed by the steering angle of the wheels, carrying out grid sampling on the external ICR of the vehicle to form external sampling meeting the requirement of simulating the movement of the vehicle rotating around a far point, and obtaining an external ICR sampling point set; According to the position coordinates of the four wheels in the vehicle parameter set, the maximum variation allowed by the steering angle of the wheels and the external ICR sampling point set, carrying out grid sampling on the internal ICR of the vehicle to form external sampling meeting the requirement of simulating the motion of the vehicle rotating around a near point, and obtaining the internal ICR sampling point set; And marking adjacent points according to the external ICR sampling point set and the internal ICR sampling point set, wherein for each pair of adjacent points, the variation of the steering angle of the wheels during transfer meets the maximum variation constraint allowed by the steering angle of the wheels in the vehicle parameter set, and an ICR grid sampling result is obtained, wherein the ICR grid sampling result comprises the coordinates of all sampling points in the external ICR sampling point set and the internal ICR sampling point set and the relation description of the adjacent ICR sampling points.
5. The path planning algorithm for a four-wheeled independent-steering vehicle of claim 4, wherein the step of raster sampling the external ICR of the vehicle based on the distances from the four wheels in the set of vehicle parameters to the center of the vehicle and the maximum amount of change allowed by the steering angle of the wheels to form an external sample that satisfies a motion simulating rotation of the vehicle about a far point, and obtaining an external ICR sample point set comprises: selecting ICR angle grid size The ICR angle grid size The calculation formula of (2) is as follows: ; Wherein, the Represents a positive integer number of the unit, Representing a maximum amount of variation permitted by the wheel steering angle; Adding ICR at infinity to a sampling grid in a polar coordinate system with the center of the vehicle as the origin, the position of the ICR sampling point being expressed as , wherein, ; Circularly executing the steps if the radius of the ICR added in the previous round is ICR radius added for this round The method comprises the following steps: ; Wherein, the Indicating the distance of the wheel from the centre of the vehicle, ICR Angle grid of Repeating the cycle until The following are not satisfied: ; Wherein, the 。
6. The path planning algorithm for a four-wheeled independent steered vehicle of claim 5, wherein said step of raster sampling the internal ICR of the vehicle from the coordinates of the positions of the four wheels in the set of vehicle parameters, the maximum amount of change allowed by the steering angle of the wheels, and the set of external ICR sampling points to form an external sample that satisfies the motion simulating the vehicle's rotation about a near point, comprises the steps of: Extracting innermost points in each direction of a vehicle according to the external ICR sampling point set, and screening out points closest to the center of the vehicle to obtain innermost point coordinates of the vehicle; According to the innermost point coordinates of the vehicle, starting from the innermost point of each direction, adding ICR sampling points to the inner circulation, and calculating new innermost point coordinates to obtain an inner ICR sampling point set, wherein the new innermost point coordinates calculate a positive half axis, a negative half axis is obtained by mirroring the positive half axis, and a positive half axis coordinate formula for calculating the new innermost point coordinates is as follows: The coordinate formula of the innermost point coordinates in the positive and negative X-axis directions is as follows: ; the coordinate formula of the innermost point coordinates in the positive and negative y-axis directions is as follows: ; Wherein, the Representing the X-coordinate of the iteratively generated new X-axis ICR sample points, Representing the X-coordinate of the current X-axis ICR sample point prior to the iteration, The reference point X coordinate of the wheel is indicated, Representing the Y-coordinate of the iteratively generated new Y-axis ICR sample points, Representing the Y-coordinate of the current Y-axis ICR sample point prior to the iteration, The reference point Y coordinate of the wheel is indicated, Indicating the amount of change in the steering angle of the wheel.
7. The path planning algorithm for a four-wheeled independently steered vehicle according to claim 4, wherein the step of obtaining the path plan for the vehicle based on the vehicle parameter set, the environmental information set, dijkstra distance values for points on the discrete map, and the ICR grid sampling result comprises: taking a starting point node of the vehicle in the environment information set as a search node, and performing heuristic cost calculation to obtain an initialized search state, wherein the initialized search state comprises an open list and a closed list, and the heuristic cost calculation formula is as follows: ; Wherein, the In order to be a heuristic cost, For the Dijkstra distance from the current point to the end point, As the weight coefficient of the light-emitting diode, For the heading angle of the target position, The current course angle; Updating the open list and the closed list by using the initialized search state, the vehicle parameter set, the environment information set and the ICR grid sampling result, selecting adjacent ICR points to generate a vehicle motion track based on the ICR grid sampling result by taking out the node with the minimum cost from the open list, performing collision detection according to the size of the vehicle and the position of the obstacle, calculating the cost of a child node, and updating the open list and the closed list to obtain a target open list and a target closed list; and taking the minimum cost as a principle, judging that the path planning fails when the target open list is empty and does not reach the destination node, otherwise, reversely traversing a father node chain from the destination node to generate a path point sequence based on the condition that the target open list is empty or reaches the destination node, and obtaining a path planning result of the vehicle.
8. A path planning system for a four-wheel independently steered vehicle, applied to the path planning algorithm for a four-wheel independently steered vehicle as defined in any one of claims 1 to 7, the system comprising: An acquisition unit, configured to acquire basic parameters of a vehicle, and determine a vehicle parameter set in a coordinate system where the vehicle is located, where the vehicle parameter set includes a size of the vehicle, position coordinates of four wheels of the vehicle, distances from the four wheels to a center of the vehicle, and a time step of path planning and a maximum variation allowed by the wheel steering angle in one time step; The system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is also used for acquiring environment information of the vehicle and obtaining an environment information set in the environment information, wherein the environment information set comprises the position of an obstacle, the boundary of a map and a starting point node and an ending point node of the vehicle; The computing unit is used for discretizing the map according to the environment information set and the vehicle parameter set, expanding the obstacle according to the size of the vehicle, and computing the distance from each position in the map to the terminal node from the terminal node in the map to obtain Dijkstra distance values of each point on the discrete map; The acquisition unit is used for carrying out grid sampling on the ICR of the vehicle according to the vehicle parameter set, forming a motion model of the vehicle meeting the limit of the change rate of the steering angle of the wheel, and obtaining an ICR grid sampling result, wherein the ICR grid sampling result comprises all sampling points and adjacent relations thereof; And the result unit is used for obtaining the path planning of the vehicle according to the vehicle parameter set, the environment information set, the Dijkstra distance value of each point on the discrete map and the ICR grid sampling result.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the algorithm of any one of claims 1 to 7 when the computer program is executed.
10. A computer storage medium having stored thereon a computer program, which when executed by a processor realizes the steps of the algorithm of any of claims 1 to 7.

Description

Path planning algorithm and system for four-wheel independent steering vehicle Technical Field The application belongs to the technical field of vehicle path planning, and particularly relates to a path planning algorithm and system for a four-wheel independent steering vehicle. Background With the rapid development of vehicle engineering and automatic driving technology, higher requirements are put on maneuvering performance and path planning capability of vehicles in narrow or complex scenes. Due to the characteristic that the steering angles of all wheels of a four-wheel independent steering (4 WIS) vehicle can be controlled independently, various maneuvering modes such as inclined travel, in-situ steering and the like can be realized, the flexibility and the pose control precision of the vehicle are obviously enhanced, and important potential is shown in application scenes such as logistics storage, special operation, unmanned operation and the like. In terms of path planning methods, currently common techniques include curve interpolation, graph search, random sampling, numerical optimization, and other intelligent algorithms. Typical planning strategies typically first generate a rough feasible path by means of graph search or random sampling, and then smooth it by optimization means. However, existing mainstream path planning methods (e.g., hybridAlgorithm) is mainly oriented to the traditional vehicle adopting the ackerman steering structure, and the special kinematic constraint and the special degree of freedom of the four-wheel independent steering vehicle cannot be fully considered. Therefore, the planned path is difficult to fully exert the omnidirectional movement potential of the four-wheel independent steering chassis, and the performance of the vehicle in a precise operation scene is restricted. Aiming at the special structure of the four-wheel independent steering vehicle, developing a high-efficiency path planning method capable of effectively integrating the multi-mode movement capability of the vehicle becomes a technical problem to be solved currently. Disclosure of Invention Based on the above, it is necessary to provide a path planning algorithm and system for four-wheel independent steering vehicles. In a first aspect, the present application provides a path planning algorithm for a four-wheel independent steering vehicle, the method comprising: Acquiring basic parameters of a vehicle, and determining a vehicle parameter set under a coordinate system of the vehicle, wherein the vehicle parameter set comprises the size of the vehicle, the position coordinates of four wheels of the vehicle, the distances from the four wheels to the center of the vehicle, a time step of path planning and the maximum variation allowed by the steering angle of the wheels in one time step; Acquiring environment information of the vehicle, and obtaining an environment information set in the environment information, wherein the environment information set comprises the position of an obstacle, the boundary of a map, and a starting point node and an end point node of the vehicle; performing discretization processing on the map according to the environment information set and the vehicle parameter set, expanding the obstacle according to the size of the vehicle, and calculating the distance from each position in the map to the terminal point from the terminal point in the map to obtain Dijkstra distance values of each point on the discrete map; According to the vehicle parameter set, carrying out grid sampling on ICR of the vehicle, and forming a motion model of the vehicle meeting the change rate limit of the steering angle of the wheel to obtain an ICR grid sampling result, wherein the ICR grid sampling result comprises all sampling points and adjacent relations thereof; and obtaining the path planning of the vehicle according to the vehicle parameter set, the environment information set, dijkstra distance values of each point on the discrete map and the ICR grid sampling result. In some embodiments, the step of obtaining the basic parameters of the vehicle and determining the vehicle parameter set under the coordinate system of the vehicle includes: acquiring the outline dimension of the vehicle and the position coordinates of the wheels, and acquiring the outline dimension of the vehicle and the position coordinates of the four wheels; calculating the distance from each wheel to the center of the vehicle by using a Euclidean distance formula according to the position coordinates of the four wheels to obtain the distance from the wheel to the center of the vehicle; And determining the time step of path planning and the maximum variation allowed by the steering angle of the wheels in one time step according to the preset motion performance requirement or constraint of the vehicle, and obtaining the maximum variation of the time step and the steering angle. In some embodiments, the discretizing the map accord