Path planning method of improved bidirectional A* algorithm

Through the improved bidirectional A* algorithm and redundant point removal strategy, the problems of low efficiency and redundancy in path planning of traditional A* algorithms are solved, and more optimized and shorter paths are generated, which improves the efficiency of robot path planning and quality.

CN119022945APending Publication Date: 2024-11-26DALIAN MARITIME UNIVERSITY
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411092855.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Traditional A* algorithms have problems such as low planning efficiency, many redundant points, unsmooth paths and excessive corners in path planning, making it difficult to find high-quality optimal paths in obstacle environments.

Method used

The improved bidirectional A* algorithm is used to combine the target bias strategy and improved heuristic function to generate initial paths through bidirectional search, and redundant point removal is taken into account the raster corner points through pruning ideas to generate paths with fewer turning points and shorter distances. .

Benefits of technology

Improve the efficiency and quality of path planning, reduce redundant points and turning points, and the generated paths are more optimized, saving planning time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119022945A_ABST
    Figure CN119022945A_ABST
Patent Text Reader

Abstract

The invention belongs to the field of mobile robot path planning, and relates to a path planning method of an improved bidirectional A * algorithm, which can obtain an optimal or suboptimal improved A * algorithm in a grid map environment. According to the algorithm, a target offset search strategy is introduced, a current node is tried to be connected with a target point when the algorithm searches a path, the search efficiency of the algorithm is improved, meanwhile, an initial path is planned by adopting a bidirectional search strategy of an improved heuristic function, and the search efficiency of the algorithm is further improved; and finally, a redundant point elimination strategy considering grid angular points is introduced, and a path with a shorter distance is further planned while the search efficiency of the algorithm is improved. According to the method, the path searching efficiency is improved, the inflection points of the generated path are reduced, and the length of the generated path is shortened.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of mobile robot path planning, and in particular to a path planning method for improving a bidirectional A* algorithm. Background Art

[0002] With the rapid development of information technology, robots are becoming more and more common in people's daily lives. Their research and development, manufacturing, and application have gradually become an important indicator of a country's scientific and technological innovation and manufacturing level, and have attracted great attention from the world's manufacturing powers. As one of the key areas vigorously promoted and developed by "Made in China 2025", the robotics industry is developing at a high speed, driving a new round of scientific and technological revolution and industrial transformation. In robot autonomous navigation, path planning is an important part of it. The result of path planning reflects the advantages and disadvantages of the motion path and ensures the safety and stability of the robot during movement. Therefore, how to plan high-quality paths is an important research direction. Robot path planning refers to the robot finding a safe and collision-free optimal path from the starting point to the target point according to the required evaluation indicators (time, distance, etc.) in an environment with obstacles.

[0003] At present, path planning algorithms can be divided into global path planning and local path planning according to the different environmental information they require. According to the algorithm principle, robot global path planning methods can be divided into search-based methods (Dijkstra, A*), sampling-based methods (PRM, RRT), bionic-based intelligent algorithms (ant colony algorithm, particle swarm algorithm, genetic algorithm), etc. Local path planning algorithms can be divided into DWA, TEB, artificial potential field method, etc. The search-based A* algorithm has the advantages of high computational efficiency and short planning path length, so it is widely used in the field of robot path planning. However, the traditional A* algorithm has problems such as low planning efficiency, many redundant points, uneven paths, and large corners. Summary of the invention

[0004] According to the technical problems mentioned in the above background technology, a path planning method of an improved bidirectional A* algorithm is provided. The present invention mainly adopts a bidirectional A* algorithm with an improved search strategy to obtain an initial path, and then performs a redundant point elimination strategy on the initial path through the pruning idea. When pruning, not only the center point of the grid is considered, but also the grid corner points are taken into consideration by judging obstacles when pruning collisions occur. While eliminating redundant path points, a path with fewer turning points and shorter distance is planned.

[0005] The technical means adopted by the present invention are as follows:

[0006] A path planning method for improving the bidirectional A* algorithm comprises the following steps:

[0007] Step 1: Scan the surrounding environment through the preset sensor equipment to establish a grid map;

[0008] Step 2: Determine the starting node of the path (map ij ) = S and the target node of the path (mapi j )=G, directly connect the starting point and the target point and perform collision detection. If there is no obstacle, the search ends, otherwise continue to step 3;

[0009] Step 3: Introduce the target bias and improved heuristic function bidirectional search strategy, and execute the improved bidirectional A* search algorithm to plan an initial path point (P);

[0010] Step 4: Execute the redundant point elimination strategy considering grid corner points on the initial path point (P) to eliminate redundant path nodes;

[0011] Step 5: Connect the paths without redundant nodes in sequence to generate the optimal path (P new ).

[0012] Furthermore, in step 1, the grid map discretizes the map environment into a number of independent square grids, each grid corresponds to a specific position in the real environment, and assigns a corresponding value to the grid according to whether there are obstacles in the real environment to indicate its idle or occupied state; all grids assigned a state together constitute a map model of the environment, and the mathematical expression is:

[0013]

[0014] Among them, M represents the entire environment map model, m and n represent the number of rows and columns corresponding to the rasterized environment information, respectively. ij Indicates the state of the grid at row i and column j, where (map ij )=0 means idle, no obstacles exist, (map ij )=1 means occupied, there is an obstacle.

[0015] Furthermore, the step 3 comprises the following steps:

[0016] Step 3.1: Initialize two open lists (openlist1), (openlist2) and two closed lists (closelist1), (closelist2) to map the starting point of the path ij )=S is put into the open list (openlist1), and the target point of the path (mapi j ) = G into openlist2;

[0017] Step 3.2: Determine whether the open lists (openlist1) and (openlist2) are empty. If they are empty, it means that the path planning fails. Otherwise, continue to step 3.3.

[0018] Step 3.3: Select the grid with the minimum cost function F(n) in the open list (openlist1) as the current node n, and delete the current node n from the open list (openlist1);

[0019] Step 3.4: Determine whether the current node n meets the target bias condition. If so, merge the current node n and the target node (mapi j )=G to connect, determine whether there is a collision with an obstacle, if there is no collision, the target node (mapi j )=G is taken as the child node of the current node n and put into the closed list (closelist1), the search is finished, and the execution continues to step 3.9, otherwise the execution continues to step 3.5;

[0020] Step 3.5: For the 8 neighboring child nodes (x n -1,y n +1)、(x n ,y n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1,y n )、( xn -1,y n -1)、(x n ,y n -1)、(x n +1,y n -1) Perform obstacle and boundary detection to determine whether the surrounding area of ​​the current node n is an obstacle (map ij )=1 or exceeds the map boundary. If it is a feasible point, ignore the nodes in the closed list (closelist1). If the domain node of the current node n is not in the open list (openlist1), add the domain node of the current node n to the open list (openlist1), and take the current node as the parent node of its domain node, and calculate G(n), H(n) and F(n) of the domain node of the current node n; if the current node n is in the open list (openlist1), if the G(n) in the domain node of the current node n is lower than the original one with the current node n as the parent node, take the current node n as the parent node, and recalculate F(n) of the domain node of the current node n, otherwise do not make any changes;

[0021] Among them, (x n ,y n ) represents the coordinate position of the current node n, then (x n -1,y n +1)、(x n ,y n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1,y n )、(x n -1,y n -1)、(x n ,y n -1)、(x n +1,y n -1) represent the 8 neighboring child nodes of the current node n respectively;

[0022] Step 3.6: Put the current node n into the closed list (closelist1);

[0023] Step 3.7: Execute steps 3.3 to 3.6 for (openlist2) and (closelist2);

[0024] Step 3.8: Determine whether the search is finished. If the search is not finished, return to step 3.3 to continue searching. Otherwise, the initial path (P) has been found, end the search, and execute step 3.9;

[0025] Step 3.9: Starting from the last node, move along each parent node, backtrack the path, and get the initial path (P).

[0026] Furthermore, in step 3.4, the target bias condition is that the number of obstacles in the eight neighborhood sub-nodes of the current node is 1 or 2.

[0027] Furthermore, in step 3, the heuristic function is:

[0028] F(n)=G(n)+(1+α)H(n)+D(n)(6);

[0029]

[0030] Where n represents the current node, G(n) represents the actual cost from the starting point of the path to the current node n, H(n) represents the estimated cost from the current node n to the target point (mapij) = G, D(n) represents the distance penalty term, which represents the vertical Euclidean distance from the current node to the straight line connecting the starting point (mapij) = S and the target point (mapij) = G, α represents the dynamic weight factor, κ represents the adjustment coefficient, (x n ,y n ) represents the coordinates of the current node in the grid map, (x n-1 ,y n-1 ) represents the coordinates of the grid map where the parent node of the current node is located, (x g ,y g ) represents the coordinates of the target point in the grid map.

[0031] Furthermore, in step 3.8, the search end condition is:

[0032] A. When the target point exists in (closelist1), the target point is the last node;

[0033] B. When the starting point exists in (closelist2), the starting point is the last node;

[0034] C. When (closelist1) and (closelist2) have the same node, the same node is the last node.

[0035] Furthermore, step 4 includes the following steps:

[0036] Step 4.1: From the target point of the path (map ij ) = G starts from the beginning of the path (map ij )=S, extract the path nodes (p1, p2…p k );

[0037] Step 4.2: Take the initial path node (p1) as the current node and put it into the optimal path list;

[0038] Step 4.3: Connect the current node (p1) to its initial path successor nodes (p2, p3…p k ) are connected in sequence for collision detection, and the last node before the collision is found (p i );

[0039] Step 4.4 Based on (p i ) and its previous and next nodes (p i-1 )、(p i+1 )Select the appropriate node;

[0040] Step 4.5 sets the selected node as the current node and puts it into the optimal path list, and repeats the above steps until the optimal path is found.

[0041] Furthermore, in step 4.4, when a collision occurs, it is determined that the obstacle that collides is located at the last node (p i ) Which direction is up, down, left or right? After determining the direction, it can be divided into four situations according to its turning angle. Select the corresponding grid corner point according to the four situations.

[0042] Compared with the prior art, the present invention has the following advantages:

[0043] 1. The present invention introduces a target bias strategy, which attempts to connect with the target point when the node is expanded. When the map environment is an open area or there are few obstacles, the node expansion efficiency is greatly improved and the planning time is saved.

[0044] 2. The present invention adopts an improved heuristic function, which reduces the number of node searches, further improves the node expansion efficiency, and saves planning time.

[0045] 3. The present invention introduces a redundant point elimination strategy to solve the problem of many redundant points and turning points in the traditional A* algorithm. When pruning, the corner points of the grid nodes are taken into consideration as path nodes to further optimize the path and reduce the length of the path. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0047] Figure 1 This is a flow chart of the path planning method for improving the bidirectional A* algorithm of the present invention.

[0048] Figure 2 Schematic diagram of the target selection strategy proposed in the present invention.

[0049] Figure 3 Schematic diagram of the grid corner point selection proposed in the present invention; (a) is a schematic diagram of the situation where the obstacle is located below and the corner point is at the upper right; (b) is a schematic diagram of the search directions of different nodes in situation (a); (c) is a schematic diagram of the situation where the obstacle is located below and the corner point is at the upper left; (d) is a schematic diagram of the search directions of different nodes in situation (c).

[0050] Figure 4Schematic diagram of the redundant point elimination strategy combined with grid corner points proposed by the present invention.

[0051] Figure 5 The figure is a schematic diagram of the optimal path planned after the specific implementation of the grid corner point removal strategy; (a) is a schematic diagram of the original path; (b) is a schematic diagram of the optimal path planned after the specific implementation of the grid corner point removal strategy. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 creative work should fall within the scope of protection of the present invention.

[0053] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0054] like Figure 1-5 As shown, the present invention provides a path planning method for improving the bidirectional A* algorithm, comprising the following steps:

[0055] Step 1: Use laser radar and other sensor equipment to scan the surrounding environment and create a grid map;

[0056] As a preferred embodiment, in the present application, step 1:

[0057] The surrounding environment information is collected by using sensors such as LiDAR to establish an environmental grid map. The grid map discretizes the map environment into several independent square grids. Each grid corresponds to a specific location in the real environment. According to whether there are obstacles in the real environment, the grid is assigned a corresponding value to indicate its idle or occupied state. All grids with assigned states together constitute the map model of the environment, and its mathematical expression is:

[0058]

[0059] Among them, M represents the entire environment map model, m and n represent the number of rows and columns corresponding to the rasterized environment information, respectively. ij Indicates the state of the grid at row i and column j, where (map ij )=0 means idle, no obstacles exist, (map ij )=1 means occupied, there is an obstacle.

[0060] Step 2: Determine the starting node of the path (map ij ) = S and the target node of the path (map ij )=G, directly connect the starting point and the target point and perform collision detection. If there is no obstacle, the search ends and execute step 3.9, otherwise continue to execute step 3;

[0061] Step 3: Introduce the target bias and improved heuristic function bidirectional search strategy, and execute the improved bidirectional A* search algorithm to plan an initial path point (P);

[0062] Further, the step 3 is specifically as follows:

[0063] Step 3.1: Initialize two open lists (openlist1), (openlist2) and two closed lists (closelist1), (closelist2), and put the starting point of the path (map ij )=S is put into the open list (openlist1), and the target point of the path (map ij ) = G into openlist2;

[0064] Step 3.2: Determine whether the open lists (openlist1) and (openlist2) are empty. If they are empty, it means that the path planning fails. Otherwise, continue to execute the following steps;

[0065] Step 3.3: Select the grid with the minimum cost function F(n) in the open list (openlist1) as the current node n, and delete the current node n from the open list (openlist1);

[0066] Step 3.4: Determine whether the current node n meets the target bias condition. If so, map the current node n to the target node (map ij )=G to connect, determine whether there is a collision with an obstacle, if there is no collision, the target node (map ij )=G is taken as the child node of the current node n and put into the closed list (closelist1), the search is finished, and the execution continues to step 3.9, otherwise the execution continues to step 3.5;

[0067] In this application, the target bias condition is that the number of obstacles in the eight neighboring child nodes of the current node is 1 or 2. Figure 2 As shown, during the node search process, N1, N2, N3, and N4 are nodes that meet the conditions and attempt to connect to the target point, and N4 successfully connects to the target point.

[0068] Step 3.5: For the surrounding area (x n -1,y n +1)、(x n ,y n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1,y n )、(x n -1,y n -1)、(x n ,y n -1)、(x n +1,y n -1) Perform obstacle and boundary detection to determine whether the surrounding area of ​​the current node n is an obstacle (map ij )=1 or exceeds the map boundary. If it is a feasible point, ignore the nodes in the closed list (closelist1). If the domain node of the current node n is not in the open list (openlist1), add the domain node of the current node n to the open list (openlist1), and take the current node as the parent node of its domain node to calculate the G(n), H(n) and F(n) of the domain node of the current node n. If the current node n is in the open list (openlist1), if the G(n) in the domain node of the current node n is lower than the original one with the current node n as the parent node, take the current node n as the parent node and recalculate the F(n) of the domain node of the current node n. Otherwise, do not make any changes. Among them, (x n ,y n ) represents the coordinate position of the current node n, then (x n -1,y n +1)、(x n ,y n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1,y n )、(x n -1,y n -1)、(xn ,y n -1)、(x n +1,y n -1) represent the 8 neighboring child nodes of the current node n respectively;

[0069] The preferred improved heuristic function of this application is:

[0070] F(n)=G(n)+(1+α)H(n)+D(n) (10);

[0071]

[0072] Among them, n represents the current node, G(n) represents the actual cost of the path from the starting point to the current node n, and H(n) represents the cost of the path from the current node n to the target point (map ij ) = estimated cost of G, D(n) represents the distance penalty term, which represents the distance from the current node to the starting point (map ij )=S and target point (map ij ) = the vertical Euclidean distance of the straight line connected to G, α represents the dynamic weight factor, κ represents the adjustment coefficient, (x n ,y n ) represents the coordinates of the current node in the grid map, (x n-1 ,y n-1 ) represents the coordinates of the grid map where the parent node of the current node is located, (x g ,y g ) represents the coordinates of the target point in the grid map.

[0073] Step 3.6: Put the current node n into the closed list (closelist1);

[0074] Step 3.7: Execute steps 3.3 to 3.6 for (openlist2) and (closelist2);

[0075] Step 3.8: Determine whether the search is finished. If the search is not finished, return to step 3.3 to continue searching. Otherwise, the initial path (P) has been found, end the search, and execute step 3.9. The search end condition is:

[0076] A. When the target point (map ij )=G exists in (closelist1), and the target point is the last node;

[0077] B. When the starting point (map ij )=S exists in (closelist2), and the starting point is the last node;

[0078] C. When (closelist1) and (closelist2) have the same node, the same node is the last node.

[0079] Step 3.9: Starting from the last node, move along each parent node, backtrack the path, and get the initial path (P).

[0080] Step 4: Execute a redundant point elimination strategy considering grid corner points on the initial path point (P) to eliminate redundant path nodes; the specific steps of step 4 are:

[0081] Step 4.1: From the target point of the path (map ij ) = G starts from the beginning of the path (map ij )=S, extract the path nodes (p1, p2…p k );

[0082] Step 4.2: Take the initial path node (p1) as the current node and put it into the optimal path list;

[0083] Step 4.3: Connect the current node (p1) with its initial path successor nodes (p2, p3…p k ) are connected in sequence for collision detection, and the last node before the collision is found (p i );

[0084] Step 4.4: Based on (p i ) and its previous and next nodes (p i-1 )、(p i+1 ) The grid corner points are selected to determine the node (p i3 ); When a collision occurs, it is determined that the obstacle that collided is located at the last node before the collision (p i ) Which direction is it, up, down, left, or right? Figure 4 The obstacle shown is located at (p i ) below, determine the direction and follow Figure 3 As shown in the figure, the rotation angle can be divided into four cases (a), (b), (c), and (d), among which (c) satisfies Figure 4 In the corresponding situation, select the corresponding grid corner point according to (c).

[0085] Step 4.5: Set the node (p i3 ) is set as the current node and put into the optimal path list, and the above steps are repeated until the optimal path is found.

[0086] Step 5: Connect the paths without redundant nodes in sequence to generate the optimal path (p new ).like Figure 5(b) is a schematic diagram of the optimal path planned after the specific implementation of the grid corner point removal strategy, where (a) is a schematic diagram of the original path.

[0087] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A path planning method for improving the bidirectional A* algorithm, characterized in that: The following steps are involved: Step 1: Scan the surrounding environment through the preset sensor equipment to establish a grid map; Step 2: Determine the starting node of the path (mapi j ) = S and the target node of the path (mapi j )=G, directly connect the starting point and the target point and perform collision detection. If there is no obstacle, the search ends, otherwise continue to step 3; Step 3: Introduce the target bias and improved heuristic function bidirectional search strategy, and execute the improved bidirectional A* search algorithm to plan an initial path point (P); Step 4: Execute the redundant point elimination strategy considering the grid corner points for the initial path point (P) to eliminate the redundant path nodes; Step 5: Connect the paths without redundant nodes in sequence to generate the optimal path (P new ).

2. The path planning method of the improved bidirectional A* algorithm according to claim 1, characterized in that: In step 1, the grid map discretizes the map environment into a number of independent square grids, each grid corresponds to a specific position in the real environment, and assigns a corresponding value to the grid according to whether there are obstacles in the real environment to indicate its idle or occupied state; all grids assigned a state together constitute the map model of the environment, and the mathematical expression is: Among them, M represents the entire environment map model, m and n represent the number of rows and columns corresponding to the rasterized environment information, respectively. ij Indicates the state of the grid at row i and column j, where (map ij )=0 means idle, no obstacles exist, (map ij )=1 means occupied, there is an obstacle.

3. The path planning method of the improved bidirectional A* algorithm according to claim 1, characterized in that: The step 3 comprises the following steps: Step 3.1: Initialize two open lists (openlist1), (openlist2) and two closed lists (closelist1), (closelist2) to map the starting point of the path ij )=S is put into the open list (openlist1), and the target point of the path (map ij ) = G into openlist2; Step 3.2: Determine whether the open lists (openlist1) and (openlist2) are empty. If they are empty, it means that the path planning fails. Otherwise, continue to step 3.

3. Step 3.3: Select the grid with the minimum cost function F(n) in the open list (openlist1) as the current node n, and delete the current node n from the open list (openlist1); Step 3.4: Determine whether the current node n meets the target bias condition. If so, map the current node n to the target node (map ij )=G to connect, determine whether there is a collision with an obstacle, if there is no collision, the target node (map ij )=G is taken as the child node of the current node n and put into the closed list (closelist1), the search is finished, and the execution continues to step 3.9, otherwise the execution continues to step 3.5; Step 3.5: For the 8 neighboring child nodes (x n -1,y n +1)、(x n ,y n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1, g n )、(x n -1,y n -1)、(x n , g n -1)、(x n +1,g n -1) Perform obstacle and boundary detection to determine whether the surrounding area of ​​the current node n is an obstacle (map ij )=1 or exceeds the map boundary. If it is a feasible point, ignore the nodes in the closed list (closelist1). If the domain node of the current node n is not in the open list (openlist1), add the domain node of the current node n to the open list (openlist1), and take the current node as the parent node of its domain node, and calculate G(n), H(n) and F(n) of the domain node of the current node n; if the current node n is in the open list (openlist1), if the G(n) in the domain node of the current node n is lower than the original one with the current node n as the parent node, take the current node n as the parent node, and recalculate F(n) of the domain node of the current node n, otherwise do not make any changes; Among them, (x n ,y n ) represents the coordinate position of the current node n, then (x n -1, g n +1)、(x n , g n +1)、(x n +1,y n +1)、(x n -1,y n )、(x n +1,g n )、(x n -1,y n -1)、(x n , g n -1)、(x n +1,y n -1) represent the 8 neighboring child nodes of the current node n respectively; Step 3.6: Put the current node n into the closed list (closelist1); Step 3.7: Execute steps 3.3 to 3.6 for (openlist2) and (closelist2); Step 3.8: Determine whether the search is finished. If the search is not finished, return to step 3.3 to continue searching. Otherwise, the initial path (P) has been found, end the search, and execute step 3.9; Step 3.9: Starting from the last node, move along each parent node, backtrack the path, and get the initial path (P).

4. The path planning method of the improved bidirectional A* algorithm according to claim 3, characterized in that: In step 3.4, the target bias condition is that the number of obstacles in the eight neighborhood sub-nodes of the current node is 1 or 2.

5. The path planning method of the improved bidirectional A* algorithm according to claim 3, characterized in that: In step 3, the heuristic function is: F(n)=G(n)+(1+α)H(n)+D(n) (2); Where n represents the current node, G(n) represents the actual cost from the starting point of the path to the current node n, and H(n) represents the cost from the current node n to the target point (map ij ) = estimated cost of G, D(n) represents the distance penalty term, which represents the distance from the current node to the starting point (map ij )=S and target point (map ij ) = the vertical Euclidean distance of the straight line connected to G, α represents the dynamic weight factor, κ represents the adjustment coefficient, (x n ,y n ) represents the coordinates of the current node in the grid map, (x n-1 , g n-1 ) represents the coordinates of the grid map where the parent node of the current node is located, (x g ,y g ) represents the coordinates of the target point in the grid map.

6. The path planning method of the improved bidirectional A* algorithm according to claim 3, characterized in that: In step 3.8, the search end condition is: A. When the target point exists in (closelist1), the target point is the last node; B. When the starting point exists in (closelist2), the starting point is the last node; C. When (closelist1) and (closelist2) have the same node, the same node is the last node.

7. The path planning method of the improved bidirectional A* algorithm according to claim 1, characterized in that: The step 4 comprises the following steps: Step 4.1: From the target point of the path (map ij ) = G starts from the beginning of the path (map ij )=S, extract the path nodes (p1, p2…p k ); Step 4.2: Take the initial path node (p1) as the current node and put it into the optimal path list; Step 4.3: Connect the current node (p1) to its initial path successor nodes (p2, p3…p k ) are connected in sequence for collision detection, and the last node before the collision is found (p i ); Step 4.4 Based on (p i ) and its previous and next nodes (p i-1 )、(p i+1 )Select the appropriate node; Step 4.5 sets the selected node as the current node and puts it into the optimal path list, and repeats the above steps until the optimal path is found.

8. The path planning method of the improved bidirectional A* algorithm according to claim 7, characterized in that: In step 4.4, when a collision occurs, it is determined that the obstacle that collides with is located at the last node before the collision occurs (p i ) Which direction is up, down, left or right? After determining the direction, it can be divided into four situations according to its turning angle. Select the corresponding grid corner point according to the four situations.

Citation Information

Cited By

  • Ground unmanned equipment cluster path planning and formation control method

    CN119311015A

  • AGV path planning method based on obstacle angular points

    CN120194719A

  • Outdoor robot path planning method based on improved bidirectional A* and DWA algorithms

    CN120506970A