Intelligent vehicle obstacle avoidance path planning method, device and storage medium

Abstract

The present invention relates to the technical field of intelligent vehicle path planning, and specifically to an intelligent vehicle obstacle avoidance path planning method, device and storable medium. The present invention combines the bidirectional A-star method with the improved artificial potential field method to perform path planning: first, a collision-free drivable path connecting the starting point and the end point is quickly planned using the bidirectional A-star method, and redundant points of the planned path are removed. The planned path is also smoothed to obtain a smooth global reference path. At the same time, in an environment where dynamic obstacles exist, the improved artificial potential field method of the present invention is used to dynamically plan the optimal local obstacle avoidance path, and move toward the target point while performing rapid dynamic obstacle avoidance. The present invention also adjusts the repulsive force used by the artificial potential field method so that it deflects when it falls into a local optimal point, thereby obtaining a resultant force for successfully escaping the local optimal point, thereby avoiding the problem that the traditional artificial potential field method is prone to falling into a local minimum.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicle path planning, and in particular to an intelligent vehicle obstacle avoidance path planning method, an intelligent vehicle obstacle avoidance path planning device using the method, and a storable medium loaded with the method. Background Art

[0002] Smart vehicles are a key area of future automotive development. They are multidisciplinary systems integrating multiple advanced technologies. A complete smart vehicle system consists of four key components: environmental perception, path planning, information processing and decision-making, and intelligent control. Within the smart vehicle sector, path planning plays a connecting role, receiving environmental information from the perception module and transmitting the planned obstacle avoidance path to the intelligent decision-making and control system.

[0003] Smart car path planning technology can be divided into global path planning and local path planning. Its main function is to find an optimal path or collision-free path from the starting point to the end point. This path is generated according to certain evaluation criteria, such as shortest distance, fastest time, safe and smooth path, etc.

[0004] The traditional A-star method is a global path planning algorithm that uses a single direction from the initial point to the target point for path planning. This results in an unsmooth and frequently inflection points. As the search space increases, the computational complexity of the A-star method increases exponentially, resulting in excessively long planning times. Furthermore, the A-star method cannot avoid dynamic obstacles. The traditional artificial potential field method is a local path planning algorithm. While this algorithm offers excellent real-time performance, rapid planning speed, and ease of implementation, it suffers from unreachable target points and a tendency to fall into local minima. Summary of the Invention

[0005] Based on this, it is necessary to provide a smart car obstacle avoidance path planning method and device to address the problem that the existing traditional A-star method alone cannot avoid dynamic obstacles.

[0006] The present invention is achieved by adopting the following technical solutions:

[0007] In a first aspect, the present invention discloses a method for intelligent vehicle obstacle avoidance path planning, comprising the following steps:

[0008] S1, obtaining an environment map of the environment in which the smart car is located;

[0009] S2, based on the environment map, establish a plane rectangular XY coordinate system, mark the starting point position coordinates, the target point position coordinates, and the obstacle position coordinates; wherein the obstacles include static obstacles and dynamic obstacles;

[0010] S3, based on the bidirectional A-star method, global path planning is performed for the smart car, and paths are searched in both forward and reverse directions to obtain an initial global path; the forward direction is from the starting point position coordinates to the target point position coordinates, and the reverse direction is from the target point position coordinates to the starting point position coordinates;

[0011] S4, remove redundant points from the initial global path to obtain a streamlined global path;

[0012] S5, smoothing the simplified global path to obtain a smoothed global reference path;

[0013] S6, based on the improved artificial potential field method, performs local path planning for the smart car, and dynamically plans a local obstacle avoidance path when dynamic obstacles appear on the smooth global reference path;

[0014] Among them, S6 includes:

[0015] Calculate the gravitational force F generated by the target point on the smart car att ; Calculate the repulsive force F generated by the obstacle on the smart car rep ; Calculate the net force F on the smart car sum , according to the net force F on the smart car sum Perform local path planning for smart cars;

[0016] Among them, when falling into the local optimal point, F rep The rotation angle θ gives the repulsive force F after rotation rep_rot , and then F rep_rot With F att Synthesize F by the parallelogram law sum .

[0017] The implementation of this intelligent vehicle obstacle avoidance path planning method is based on the method or process of the embodiment of the present disclosure.

[0018] The present invention firstly uses a bidirectional A-star method to quickly plan a collision-free drivable path connecting a starting point and an end point, removes redundant points of the planned path, and smoothes the planned path to obtain a smooth global reference path; at the same time, in an environment where dynamic obstacles exist, the improved artificial potential field method of the present invention is used to dynamically plan an optimal local obstacle avoidance path, and moves toward a target point while performing rapid dynamic obstacle avoidance; the present invention further adjusts the repulsive force used by the improved artificial potential field method so that it deflects when trapped in a local optimal point, thereby obtaining a resultant force for successfully escaping the local optimal point, thereby avoiding the problem that the traditional artificial potential field method is prone to falling into a local minimum.

[0019] In a second aspect, the present invention discloses an intelligent vehicle obstacle avoidance path planning device, which uses the intelligent vehicle obstacle avoidance path planning method disclosed in the first aspect.

[0020] The intelligent vehicle obstacle avoidance path planning device includes: an environment map acquisition module, a coordinate marking module, a global path planning module, a global path simplification module, a global path smoothing module, and a local path planning module.

[0021] The environmental map acquisition module is used to obtain the environmental map of the environment in which the smart car is located. The coordinate marking module is used to establish a plane rectangular XY coordinate system based on the environmental map, and mark the starting position coordinates, target point position coordinates, and obstacle position coordinates of the smart car. The global path planning module is used to perform global path planning for the smart car based on the bidirectional A-star method, and simultaneously search for paths in the forward and reverse directions to obtain an initial global path. The global path simplification module is used to remove redundant points from the initial global path to obtain a simplified global path. The global path smoothing module is used to smooth the simplified global path to obtain a smoothed global reference path. The local path planning module is used to perform local path planning for the smart car based on the improved artificial potential field method, and dynamically plan a local obstacle avoidance path when a dynamic obstacle appears on the smooth global reference path.

[0022] The implementation of the intelligent vehicle obstacle avoidance path planning device is based on the method or process of the embodiment of the present disclosure.

[0023] In a third aspect, the present invention discloses a readable storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the intelligent vehicle obstacle avoidance path planning method disclosed in the first aspect is executed.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] The present invention can ensure the real-time, smoothness and safety of the planned path, and is more in line with practical applications. The present invention first uses a bidirectional A-star method to quickly plan a collision-free drivable path connecting a starting point and an end point, removes redundant points of the planned path, and smoothes the planned path to obtain a smooth global reference path. At the same time, in an environment where dynamic obstacles exist, the improved artificial potential field method of the present invention is used to dynamically plan an optimal local obstacle avoidance path, and move toward a target point while performing rapid dynamic obstacle avoidance. The present invention further adjusts the repulsive force used by the improved artificial potential field method so that it deflects when trapped in a local optimal point, thereby obtaining a resultant force for successfully escaping the local optimal point, thereby avoiding the problem that the traditional artificial potential field method is prone to falling into a local minimum. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] Figure 1 This is a brief flowchart of the intelligent vehicle obstacle avoidance path planning method proposed in Example 1 of the present invention;

[0028] Figure 2 This is a detailed flow chart of the intelligent vehicle obstacle avoidance path planning method proposed in Example 1 of the present invention;

[0029] Figure 3 This is a schematic diagram showing the principle of the intelligent vehicle obstacle avoidance path planning method proposed in Example 1 of the present invention for solving the problem that the traditional artificial potential field method is prone to falling into the local optimum point;

[0030] Figure 4 This is a simulation diagram of the global reference path planned using the traditional A-star method in Example 3 of the present invention;

[0031] Figure 5 This is an initial global path simulation diagram planned using the bidirectional A-star method in Example 3 of the present invention;

[0032] Figure 6 For Figure 5 A simplified global path simulation diagram after removing redundant points from the initial global path;

[0033] Figure 7 For Figure 6 A smoothed global reference path simulation diagram after smoothing the simplified global path;

[0034] Figure 8 This is a path simulation diagram planned using the traditional artificial potential field method in Example 3 of the present invention;

[0035] Figure 9 This is a complete path simulation diagram of embodiment 3 of the present invention using the intelligent vehicle obstacle avoidance path planning method of embodiment 1 to avoid obstacles in the presence of both dynamic and static obstacles. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] It should be noted that when a component is referred to as being "mounted on" another component, it may be directly on the other component or there may be a central component. When a component is considered to be "set on" another component, it may be directly set on the other component or there may be a central component. When a component is considered to be "fixed to" another component, it may be directly fixed to the other component or there may be a central component.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "or / and" as used herein includes any and all combinations of one or more of the associated listed items.

[0039] Example 1

[0040] This embodiment 1 aims to provide a method for planning a path to avoid obstacles for a smart car. Figure 1 , is a brief flow chart of the intelligent vehicle obstacle avoidance path planning method. Figure 2 , which is a detailed flow chart of the intelligent vehicle obstacle avoidance path planning method.

[0041] A method for intelligent vehicle obstacle avoidance path planning includes the following steps:

[0042] S1, obtaining an environment map of the environment in which the smart car is located.

[0043] The first step is to obtain information about the environment that the smart car is in. Generally, existing map information can be used as a basis, such as map software that provides real-time road conditions.

[0044] S2. Based on the environment map, establish a plane rectangular XY coordinate system, mark the starting point position coordinates, the target point position coordinates, and the obstacle position coordinates; wherein the obstacles include static obstacles and dynamic obstacles.

[0045] The coordinates of this method are divided into grids, similar to taking pixels. Specifically, S2 includes:

[0046] Establish a plane rectangular XY coordinate system in the environment map;

[0047] Divide the grid map into a rectangular XY coordinate system;

[0048] The starting point position, target point position, and obstacle position are represented by the grid coordinates of the grid map.

[0049] Generally, the starting point coordinates are set as (x start ,y start), the target point position coordinates are set to (x final ,y final ).

[0050] The obstacle position coordinates are set as (x obstacle ,y obstacle ), which is the general expression of the coordinates of dynamic and static obstacles. Among them: the coordinate expression of static obstacles is (x 静 ,y 静 );

[0051] The coordinate representation of dynamic obstacles is (x 动 ,y 动 ,X 动 ,Y 动 ,T 动 ). (x 动 ,y 动 ) represents the coordinates of the center of the circle where the dynamic obstacle starts moving; X 动 Indicates movement in the horizontal direction. A positive number indicates movement in the positive direction of the horizontal axis, and a negative number indicates movement in the negative direction of the horizontal axis. 动 Indicates movement in the direction perpendicular to the horizontal axis. A positive number indicates movement in the positive direction perpendicular to the horizontal axis, and a negative number indicates movement in the negative direction perpendicular to the horizontal axis. 动 Indicates the period of repeated movement of dynamic obstacles. The larger the value, the slower the movement.

[0052] It should be noted that if the obstacle is a static obstacle, x obstacle Take x 静 ,y obstacle Take y 静 ; If the obstacle is a dynamic obstacle, x obstacle Take x 动 ,,y obstacle Take y 动 .

[0053] In addition, the current position coordinates of the smart car are set as (x now ,y now ).

[0054] S3, based on the bidirectional A-star method, performs global path planning for the smart car, searches for paths in both forward and reverse directions simultaneously, and obtains the initial global path; wherein the forward direction is from the starting point position coordinates to the target point position coordinates, and the reverse direction is from the target point position coordinates to the starting point position coordinates.

[0055] Since the paths obtained by the traditional A-star method often have many redundant points and turning points, and the search time is also long, the present invention adopts a bidirectional A-star method to search for the path from the starting point position coordinates and the target point position coordinates at the same time - during the search process, the nodes generated in one direction are used as the target nodes in the other direction.

[0056] Specifically, S3 includes:

[0057] Initialize the forward and reverse open lists and the forward and reverse close lists;

[0058] Put the coordinates of the starting position and the target position into the forward and reverse open lists, and clear the forward and reverse close lists;

[0059] Traverse each point in the forward and reverse open lists, calculate the cost value in turn, and select the point with the smallest cost value, remove it from the forward and reverse open lists, and add it to the forward and reverse close lists;

[0060] Set the current position of the forward search to S and use it as the target node for the reverse search; set the current position of the reverse search to G2 and use it as the target node for the forward search;

[0061] Perform path search in both forward and reverse directions simultaneously;

[0062] Until the current position of the reverse search and the current position of the forward search coincide with the same target node, and the target node meets the constraints of the forward and reverse searches, the initial global path is obtained.

[0063] Since the forward and reverse search processes are performed simultaneously, the number of search points is effectively reduced, improving search efficiency. The initial global path obtained is significantly shorter than the traditional A-star method, but there are still cases where there are too many redundant points in the path and the path is not smooth.

[0064] S4, remove redundant points from the initial global path to obtain a streamlined global path.

[0065] In summary, the S4 includes:

[0066] Select M reference points at intervals on the initial global path; the starting point of the path is the first reference point, and the end point of the path is the Mth reference point;

[0067] Connect the Ath reference point with the subsequent MA reference points in sequence until the line between the Ath reference point and the Bth reference point is covered by a static obstacle;

[0068] The line between the Ath reference point and the B-1th reference point is used as a simplified path segment;

[0069] A ranges from 1 to M to obtain a simplified global path, and M is determined based on the actual situation.

[0070] In other words, the starting point of the initial global path is recorded as the first reference point, and then M-1 reference points are set at intervals along the initial global path in the forward direction, and the end point of the path is taken as the Mth reference point; the first reference point is connected to the second reference point. If the line between the two is not covered by a static obstacle, the second reference point is no longer considered; the first reference point is connected to the third reference point. If the line between the two is not covered by a static obstacle, the third reference point is no longer considered; the points are connected in sequence until a static obstacle is covered between the two points; if the line between the first reference point and the fourth reference point is covered by a static obstacle, the line between the first reference point and the third reference point is regarded as a successfully pruned path; the third reference point is connected to the subsequent reference points in sequence, and the above steps are repeated to finally obtain a streamlined global path after removing redundant points.

[0071] S5, smoothing the simplified global path to obtain a smoothed global reference path.

[0072] Although the simplified global path is shorter, it is still not smooth. Therefore, S5 includes: using a quasi-uniform cubic B-spline curve to smooth the curvature discontinuities in the simplified global path to obtain a smooth global reference path.

[0073] Generally speaking, the higher the degree, the higher the derivative degree of the curve, which will result in more zero points. More derivative zero points will lead to more extreme values in the original curve, causing the curve to have more peaks and valleys. The lower the degree, the better the spline curve approximates the control points. On the other hand, the cubic B-spline curve can achieve second-order derivative continuity. Therefore, the quasi-uniform cubic B-spline curve is selected for smoothing.

[0074] S6, based on the improved artificial potential field method, performs local path planning for the smart car, and dynamically plans a local obstacle avoidance path when dynamic obstacles appear on the smooth global reference path;

[0075] If there are dynamic obstacles in the environment where the smart car is located, dynamic obstacle avoidance is required.

[0076] S6 includes:

[0077] Calculate the gravitational force F generated by the target point on the smart car att ; Calculate the repulsive force F generated by the obstacle on the smart car rep ; Calculate the net force F on the smart car sum , according to the net force F on the smart car sum Perform local path planning for smart cars.

[0078] Among them, F att Including the X-axis gravity F att_x 、Fatt_y ;

[0079] in,

[0080]

[0081] Where K att represents the gravitational potential field gain coefficient, x final Indicates the horizontal coordinate of the target point position, y final Indicates the vertical coordinate of the target point, x now Indicates the horizontal coordinate of the current position of the smart car, y now represents the vertical coordinate of the current position of the smart car, and α represents the weight coefficient.

[0082] It should be noted that simulation experiments have verified that when α is 0.75, the planned path is safer, and can improve the path-finding efficiency and the success rate of reaching the target point, and reduce the planning time.

[0083] F rep Including the static obstacle repulsion F in the X-axis direction rep_x1 , Y-axis static obstacle repulsion F rep_y1 , X-axis direction dynamic obstacle repulsion F rep_x2 , dynamic obstacle repulsion F in the Y-axis direction rep_y2 ;

[0084]

[0085]

[0086]

[0087]

[0088] Where K rep represents the repulsive potential field gain coefficient, x obstacle Indicates the horizontal coordinate of the obstacle position, y obstacle Indicates the vertical coordinate of the obstacle position,

[0089] x now Indicates the horizontal coordinate of the current position of the smart car, y now represents the vertical coordinate of the current position of the smart car, β represents the angle between the current position and the obstacle, d 01 Indicates the distance affected by the static obstacle potential field, d 02 Indicates the influence distance of the dynamic obstacle potential field;

[0090] Among them, if the distance between the smart car and the static obstacle is greater than d 01 , the smart car is not affected by the repulsive force of static obstacles, that is, F rep_x1 、Frep_y1 All are reset to zero;

[0091] If the distance between the smart car and the dynamic obstacle is greater than d 02 , the smart car is not affected by the repulsive force of dynamic obstacles, that is, F rep_x2 、F rep_y2 All reset to zero.

[0092] F sum Including the resultant force F in the X-axis direction sum_x , the resultant force F in the Y-axis direction sum_y ;

[0093] Among them, when not falling into the local optimal point,

[0094] F sum_x =F att_x -F rep_x1 -F rep_x2 ;

[0095] F sum_y =F att_y -F rep_y1 -F rep_y2 .

[0096] However, if the smart car falls into the local optimal point, that is, the smart car is subject to F in real time att With F rep For the convenience of explanation, Figure 3 For example: Figure 3 The smart car shown (i.e. Figure 3 O) toward the target position (i.e. Figure 3 Goal) moves and is affected by gravity F att , and is also subject to two obstacles (i.e. Figure 3 The repulsive force F of A and B rep1 、F rep2 , where F rep1 is the repulsive force between A and O, F rep2 is the repulsive force between B and O; F rep1 and F rep2 The net force is F rep and with F att Collinear.

[0097] At this time, F rep The rotation angle θ gives the repulsive force F after rotation rep_rot , and then F rep_rot With F att Synthesize F by the parallelogram law sum , which enables the smart car to successfully escape from the local optimum. It should be noted that after simulation experiments, the effect is better when θ is 90°.

[0098] Example 2

[0099] This embodiment 2 discloses an intelligent vehicle obstacle avoidance path planning device, which uses the intelligent vehicle obstacle avoidance path planning method of embodiment 1.

[0100] The intelligent vehicle obstacle avoidance path planning device includes an environment map acquisition module, a coordinate marking module, a global path planning module, a global path simplification module, a global path smoothing module, and a local path planning module.

[0101] The environmental map acquisition module is used to obtain the environmental map of the environment in which the smart car is located. The coordinate marking module is used to establish a plane rectangular XY coordinate system based on the environmental map, and mark the starting position coordinates, target point position coordinates, and obstacle position coordinates of the smart car. The global path planning module is used to perform global path planning for the smart car based on the bidirectional A-star method, and simultaneously search for paths in the forward and reverse directions to obtain an initial global path. The global path simplification module is used to remove redundant points from the initial global path to obtain a simplified global path. The global path smoothing module is used to smooth the simplified global path to obtain a smoothed global reference path. The local path planning module is used to perform local path planning for the smart car based on the improved artificial potential field method, and dynamically plan a local obstacle avoidance path when a dynamic obstacle appears on the smooth global reference path.

[0102] This embodiment 2 also discloses a readable storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the intelligent vehicle obstacle avoidance path planning method disclosed in embodiment 1 is executed.

[0103] When the method of Example 1 is applied, it can be applied in the form of software, such as a program designed as a computer-readable storage medium that can run independently. The computer-readable storage medium can be a USB flash drive designed as a USB shield, and the USB flash drive is designed to start the program of the entire method through external triggering.

[0104] Example 3

[0105] The inventors conducted simulation verification of the intelligent vehicle obstacle avoidance path planning method of Example 1. They used Matlab to build a plane rectangular XY coordinate system, in which a grid map was divided into 20*20 grids, i.e., the length and width were set to 20 squares respectively. They also customized obstacles.

[0106] First, only static obstacles are set. Among them, there are 22 static obstacles, which are set in the grid map at: row 4 to row 7, column 5 to column 7, row 11 to row 12, column 8 to column 10, row 16 to row 17, column 10 to column 11, row 20, column 13 to column 14, row 15 to row 16, column 15 to column 16. The specific coordinates are "(4.5, 4.5), (5.5, 4.5), (6.5, 4.5), (4.5, 5.5), (5.5, 5.5), (6.5, 5.5), (12.5, 8.5), (13.5, 8.5), (12.5, 9.5), (13.5, 9.5), (7.5, 10.5), (8.5, 10.5), (9.5, 10.5), (7.5, 11.5), (8.5, 11.5), (9.5, 11.5), (9.5, 15.5), (10.5, 15.5), (9.5, 16.5), (10.5, 16.5), (12.5, 19.5), (13.5, 19.5)".

[0107] See Figure 4 ,Using the traditional A-star method to plan the global reference path, it can be seen that it has many redundant points and turning points, ,which illustrates the disadvantage of using the traditional A-star method alone.

[0108] See Figure 5 , the initial global path planned using the bidirectional A-star method. Compared to Figure 4 , Figure 5 The path is significantly shortened, but there are still problems such as too many redundant points and an uneven path. Figure 6 ,right Figure 5 The initial global path is processed to remove redundant points. Figure 5 , Figure 6 The path length is shortened, but the problem of uneven path still exists. Figure 7 ,right Figure 6 The simplified global path is smoothed. Compared with Figure 6 , Figure 7 The smoothed global path is smoother.

[0109] Also, see Figure 8 For comparison, a static obstacle was placed in another situation, and the traditional artificial potential field method was used for local path planning. The resulting path was too close to the static obstacle, making collisions likely, which also illustrates the disadvantage of using the traditional artificial potential field method alone.

[0110] Then, in Figure 7Two dynamic obstacles have been added to the existing obstacle settings. Their specific coordinates on the grid map are [16, 14, -0.01, 0.01, 80] and [8.5, 7.5, -0.03, -0.03, 40]. For example, [16, 14, -0.01, 0.01, 80] indicates that the starting point of this dynamic obstacle is at position 16 on the X-axis and 14 on the Y-axis. A negative value of -0.01 indicates movement in the negative direction of the horizontal axis, while a positive value of 0.01 indicates movement in a direction perpendicular to the horizontal axis and in the positive direction perpendicular to the horizontal axis. 80 represents the period of this dynamic obstacle's repeated movement.

[0111] See Figure 9 , the path planning is performed using the smart car obstacle avoidance path planning method of Example 1. The solid line path is the complete path obtained by the smart car according to the method of Example 1, and the dotted line path is Figure 7 Same smooth global path.

[0112] in accordance with Figure 9 It can be seen that the smart car moves from the starting position to the target position along the solid line path, passing through two dynamic obstacles and completing dynamic obstacle avoidance, which shows that the smart car obstacle avoidance path planning method of Example 1 is effective and feasible.

[0113] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0114] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A method for intelligent vehicle obstacle avoidance path planning, characterized in that: The following steps are involved: S1, obtaining an environment map of the environment in which the smart car is located; S2, based on the environment map, establish a plane rectangular XY coordinate system, mark the starting point position coordinates, the target point position coordinates, and the obstacle position coordinates; wherein the obstacles include static obstacles and dynamic obstacles; S3, based on the bidirectional A-star method, global path planning is performed for the smart car, and paths are searched in both forward and reverse directions to obtain an initial global path; the forward direction is from the starting point position coordinates to the target point position coordinates, and the reverse direction is from the target point position coordinates to the starting point position coordinates; S4, remove redundant points from the initial global path to obtain a streamlined global path; S5, smoothing the simplified global path to obtain a smoothed global reference path; S6, based on the improved artificial potential field method, performs local path planning for the smart car, and dynamically plans a local obstacle avoidance path when dynamic obstacles appear on the smooth global reference path; Among them, S6 includes: Calculate the gravitational force F generated by the target point on the smart car att ; Calculate the repulsive force F generated by the obstacle on the smart car rep ; Calculate the net force F on the smart car sum , according to the net force F on the smart car sum Perform local path planning for smart cars; Among them, when falling into the local optimal point, F rep The rotation angle θ gives the repulsive force F after rotation rep_rot , and then F rep_rot With F att Synthesize F by the parallelogram law sum .

2. The intelligent vehicle obstacle avoidance path planning method according to claim 1, characterized in that S2 include: Establish a plane rectangular XY coordinate system in the environment map; Divide the grid map into a rectangular XY coordinate system; The starting point position, target point position, and obstacle position are represented by the grid coordinates of the grid map.

3. The intelligent vehicle obstacle avoidance path planning method according to claim 2, characterized in that: S3 includes: Initialize the forward and reverse open lists and the forward and reverse close lists; Put the coordinates of the starting position and the target position into the forward and reverse open lists, and clear the forward and reverse close lists; Traverse each point in the forward and reverse open lists, calculate the cost value in turn, and select the point with the smallest cost value, remove it from the forward and reverse open lists, and add it to the forward and reverse close lists; Set the current position of the forward search to S and use it as the target node for the reverse search; set the current position of the reverse search to G2 and use it as the target node for the forward search; Perform path search in both forward and reverse directions simultaneously; Until the current position of the reverse search and the current position of the forward search coincide with the same target node, and the target node meets the constraints of the forward and reverse searches, the initial global path is obtained.

4. The intelligent vehicle obstacle avoidance path planning method according to claim 3, characterized in that S4 include: Select M reference points at intervals on the initial global path; the starting point of the path is the first reference point, and the end point of the path is the Mth reference point; Connect the Ath reference point with the subsequent MA reference points in sequence until the line between the Ath reference point and the Bth reference point is covered by a static obstacle; The line between the Ath reference point and the B-1th reference point is used as a simplified path segment; Among them, A takes values from 1 to M to obtain a simplified global path.

5. The intelligent vehicle obstacle avoidance path planning method according to claim 4, characterized in that S5 include: The curvature discontinuity points in the simplified global path are smoothed using a quasi-uniform cubic B-spline curve to obtain a smooth global reference path.

6. The intelligent vehicle obstacle avoidance path planning method according to claim 1, characterized in that: In S6, F att Including the X-axis gravity F att_x 、F att_y ; in, Among them, K att represents the gravitational potential field gain coefficient, x final Indicates the horizontal coordinate of the target point position, y final Indicates the vertical coordinate of the target point, x now Indicates the horizontal coordinate of the current position of the smart car, y now represents the vertical coordinate of the current position of the smart car, and α represents the weight coefficient.

7. The intelligent vehicle obstacle avoidance path planning method according to claim 1 or 6, characterized in that: In S6, F rep Including the static obstacle repulsion F in the X-axis direction rep_x1 , static obstacle repulsion F in the Y-axis direction rep_y1 , X-axis direction dynamic obstacle repulsion F rep_x2 , dynamic obstacle repulsion F in the Y-axis direction rep_y2 ; Among them, K rep represents the repulsive potential field gain coefficient, x obstacle Indicates the horizontal coordinate of the obstacle position, y obstacle Indicates the vertical coordinate of the obstacle position, x now Indicates the horizontal coordinate of the current position of the smart car, y now represents the vertical coordinate of the current position of the smart car, β represents the angle between the current position and the obstacle, d 01 Indicates the distance affected by the static obstacle potential field, d 02 Indicates the influence distance of the dynamic obstacle potential field; Among them, if the distance between the smart car and the static obstacle is greater than d 01 , F rep_x1 、F rep_y1 All are reset to zero; If the distance between the smart car and the dynamic obstacle is greater than d 02 , F rep_x2 、F rep_y2 All reset to zero.

8. The intelligent vehicle obstacle avoidance path planning method according to claim 7, characterized in that: In S6, F sum Including the resultant force F in the X-axis direction sum_x , the resultant force F in the Y-axis direction sum_y ; Among them, when not falling into the local optimal point, F sum_x =F att_x -F rep_x1 -F rep_x2 ; F sum_y =F att_y -F rep_y1 -F rep_y2 。 9. An intelligent vehicle obstacle avoidance path planning device, characterized in that: Using the intelligent vehicle obstacle avoidance path planning method according to any one of claims 1 to 8; The intelligent vehicle obstacle avoidance path planning device comprises: An environment map acquisition module, which is used to obtain an environment map of the environment in which the smart car is located; A coordinate marking module is used to establish a plane rectangular XY coordinate system based on the environment map, marking the coordinates of the starting point position, target point position, and obstacle position of the smart car; The global path planning module is used to plan the global path of the smart car based on the bidirectional A-star method, searching for paths in both forward and reverse directions to obtain the initial global path; A global path simplification module is used to remove redundant points from the initial global path to obtain a simplified global path; a global path smoothing module, configured to smooth the simplified global path to obtain a smoothed global reference path; and The local path planning module is used to perform local path planning for the intelligent vehicle based on the improved artificial potential field method, and dynamically plan a local obstacle avoidance path when dynamic obstacles appear on the smooth global reference path.

10. A readable storage medium, characterized in that: The readable storage medium stores computer program instructions, and when the computer program instructions are read and executed by a processor, the intelligent vehicle obstacle avoidance path planning method according to claims 1-8 is executed.

Images