Robot path planning optimization method based on improved A-STAR algorithm

Abstract

The invention discloses a robot path planning optimization method based on an improved A-STAR algorithm, and relates to the field of robots. Comprising the following steps: performing target position node initialization by using an A-STAR path planning algorithm; checking the positions of adjacent grid nodes, and determining a target node # imgabs0 #; searching is carried out by using an optimization-based RMB; selecting and updating the node with the lowest total cost by using a self-adaptive cost function; and selecting the updated node with the lowest total cost, and determining the optimal path planning. By combining the adaptive cost function and the optimized RMB, the target of remarkably improving the path planning efficiency and speed on the premise of keeping the path cost reasonable is achieved; according to the method, the search range can be adjusted according to specific conditions in the search process, unnecessary search nodes are reduced, and therefore time complexity is reduced, and path planning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of robots, and in particular to a robot path planning optimization method based on an improved A-STAR algorithm. Background Art

[0002] In the field of robotics, path planning is a critical task that determines how to reach the optimal target location with the lowest cost. In this research field, the main goal is to develop efficient path planning algorithms and improve their performance in terms of search time complexity, path cost, and search area. Current robot path planning algorithms include global path planning and local path planning. Robots using global path planning infer the optimal path based on overall environmental information. Compared with local path planning, which searches for paths in unknown environments, it is more time-saving and efficient. In the field of global path planning, the graph-based A-STAR search algorithm is widely used. This method constructs a path tree starting from the starting node and adds nodes as it searches the graph until the target node is reached or all options are exhausted. The final result is the shortest path found in the search path tree.

[0003] It has the following shortcomings: it may not achieve optimal performance when dealing with complex environments and tasks, and the strategy of finding the shortest path in the search path tree needs to be further optimized.

[0004] To this end, the present invention proposes a robot path planning optimization method based on an improved A-STAR algorithm. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a robot path planning optimization method based on an improved A-STAR algorithm.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: The robot path planning optimization method based on the improved A-STAR algorithm includes the following steps: S1: Use the A-STAR path planning algorithm to initialize the target location node; S2: Check the position of adjacent grid nodes and determine the target node ; S3: Search using optimized RMB; S4: Using the adaptive cost function, select the node with the lowest total cost and update it; S5: Select the updated node with the lowest total cost and determine the optimal path planning.

[0007] Preferably, the step S1 includes the following steps: S11: First, initialize the starting node to a time node with a cost of 0; S12: Add the starting node to the OPEN list, which is used to store nodes to be expanded.

[0008] Preferably, the step S2 includes the following steps: S21: Search surrounding nodes on the weighted graph; S22: If the surrounding nodes are obstacle nodes , then ignore the node; S23: If the surrounding nodes are not obstacle nodes, add them to the OPEN list; S24: When the list is not empty, select the node with the lowest total cost and delete it from the OPEN list.

[0009] Preferably: in the step S24, the cost function is Calculate the node cost, where Represents a heuristic function that uses the Euclidean distance.

[0010] Preferably, the step S24 includes the following steps: S241: Search and generate all possible successor nodes of the selected node, and calculate their costs and heuristic values according to the cost function; S243: If the selected node is the target node, stop searching and return the optimal path; S244: For each successor node, if it has not been visited or has a higher cost, update its cost and heuristic value; S245: If each updated successor node does not already appear in the list, add it to the list; S246: If the successor node is already in the list and the new value is lower than the previous value, update its cost and heuristic value; if the successor node is already in the visit list, ignore the node; S247: Repeat the above steps until every node on the searched weighted grid graph is traversed.

[0011] Preferably, in the step S3, the robot uses a unit block (RMB) to search for adjacent nodes, and the unit block consists of eight units surrounding the robot.

[0012] Preferably: in the step S3, the unit block is represented as ;in, Indicates the size of the motion block, Indicates the cost of movement in the four directions of up, down, left, and right; Represents the movement cost in the four diagonal directions; represents the cost associated with the four basic movements in the directions of up, down, left, and right, represents the costs associated with movement in the four diagonal directions of up-right, up-left, down-right, and down-left; represents the adaptive cost.

[0013] Preferably: in step S4, the adaptive cost function is ,in, Represents the current node, represent The cost of moving represents the motion matrix of the node, represents the target node, Used to increase the importance of the distance value between each motion matrix node and the target node; in order to determine the motion matrix node and The cost of the , formulated an adaptive cost function , which multiplies the basic movement and diagonal Costs associated with exercise.

[0014] Preferably, the step S4 includes the following steps: S41: First, the cost of the current node With the current node and Motion Matrix nodes Add the Euclidean distance between them to get the absolute difference between the x-coordinate and the y-coordinate; S42: With the motion matrix Nodes and target nodes Add the Euclidean distances between them and multiply by a constant ; In the step S42, Used to increase the importance of the distance value between each node of the motion matrix and the target node.

[0015] Preferably: in the step S4, in order to adapt to the different output results of the motion blocks caused by the different motion blocks of the robot, the method for obtaining the optimal motion block adopts the formula ,in, Represents An array of elements, each element contains the average value of the output data, according to the size of RMB, in the data set containing data points Conduct experiments in Indicates parameters, Represents an element of a single array output of size 1 per QMB, where ; The specific method is: A1: Use the formula to equally weight each data matrix and put it into the range of R=1000 to obtain a range array. ; A2: Sum each index data point in all result data matrices and merge them by the total number of arrays to obtain ; A3: Determine the best by selecting the minimum value of the final result data array for each RMB .

[0016] The beneficial effects of the present invention are: 1. By combining the adaptive cost function and optimized RMB, the present invention achieves the goal of significantly improving the efficiency and speed of path planning while keeping the path cost reasonable.

[0017] 2. The present invention can adjust the search range according to the specific circumstances during the search process, reducing unnecessary search nodes, thereby reducing time complexity and improving the efficiency of path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 Schematic diagram of the framework of the robot path planning optimization method based on the improved A-STAR algorithm proposed in the present invention; Figure 2 This is a flow chart of the robot path planning optimization method based on the improved A-STAR algorithm proposed in the present invention; Figure 3 This is a schematic diagram of the path planning of the robot path planning optimization method based on the improved A-STAR algorithm proposed in this invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be further described in detail below in conjunction with specific implementation methods.

[0020] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "connected," "connected," and "disposed" should be understood in a broad sense. For example, they may refer to fixed connection or disposition, detachable connection or disposition, or integral connection or disposition. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances. Example 1:

[0021] The robot path planning optimization method based on the improved A-STAR algorithm includes the following main steps. The overall process is as follows: Figure 2 As shown: S1. First, start the A-STAR path planning algorithm and initialize the target location node. Specifically, the A-STAR algorithm, as a best-first search or informed search algorithm, operates on a weighted graph, with the goal of finding the target node with the minimum travel cost and time starting from a specified starting node. Using the optimized A-STAR algorithm proposed in this invention, the starting node (Start) is first initialized to a node with a cost of 0 and added to the OPEN list, which is used to store the nodes to be expanded. S2. Secondly, check the position of the adjacent grid nodes and determine the target node ; Specifically, the proposed Robotic Motion Block (RMB) is used to check the neighbor nodes of the current node. If the neighbor node is an obstacle node ( ), ignore it and only add the terminal node to the list; otherwise, add it to the OPEN list. When the list is not empty, select the node with the lowest total cost and delete it from the list. The calculation of the cost function is as follows: ; in, Represents a heuristic function using Euclidean distance; if the selected node is the target node, stop the search and return the optimal path. Generate all possible successor nodes of the selected node and calculate their costs and heuristic values. For each successor node, if it has not been visited or has a higher cost, update its cost and heuristic value. Add each updated successor node to the list if it does not already appear in the list. If the successor node is already in the list, update its cost and heuristic value, provided that the new value is lower than the previous value. If the successor node is already in the visited list (i.e., it has been expanded), ignore the node. If all nodes in the motion block unit have been checked, repeat the above steps until every node on the grid graph has been checked. Terminate the algorithm when the target node is found or the list is empty; S3. Search using optimized RMB. In a grid-based search, the robot must search the surrounding cells to find the target node within as few cells as possible. The robot uses a cell block (robotmotionblock, RMB) to search its neighboring nodes. However, the traditional RMB used by the ordinary A-STARs algorithm results in an increase in the number of search cells and a longer search time. The motion block proposed in this invention surrounds the robot with eight neighboring cells, aiming to minimize the number of search cells and the time required to find the target location while maintaining a comparable path cost. Specifically, RMB is represented as follows: ; in, Indicates the size of the motion block, Indicates the cost of movement in the four directions of up, down, left, and right; Represents the movement cost in the four diagonal directions; represents the cost associated with the four basic movements in the directions of up, down, left, and right, represents the costs associated with movement in the four diagonal directions of up-right, up-left, down-right, and down-left; represents the adaptive cost, Figure 3 The figure shows the schematic diagram of the optimized RMB proposed in the present invention, where the arrows represent the search directions and the dots represent the search nodes; S4. Based on the proposed adaptive cost function, the node with the lowest total cost is selected and updated. The optimized RMB has eight adjacent unit directions, including up, down, left, right, up-right, up-left, down-right, and down-left. The costs of the four positive direction points and the diagonal direction points are respectively expressed as and These costs are calculated using the following formula: ;in, Represents the current node, represent The cost of moving represents the motion matrix of the node, represents the target node, Used to increase the importance of the distance value between each motion matrix node and the target node. and The cost of the , formulated an adaptive cost function , which multiplies the basic movement and diagonal Movement-related costs. In this adaptive cost function, the cost of the current node is first With the current node and Motion Matrix nodes The Euclidean distance between them is added to get the absolute difference between the x-coordinate and the y-coordinate. Determine as shown in the formula. Then, With the motion matrix Nodes and target nodes Add the Euclidean distances between them and multiply by a constant .here, It is used to increase the importance of the distance value between each node of the motion matrix and the target node. Since the robot motion blocks have different sizes and the output results of each motion block are also different. In order to obtain the optimal RMB, the present invention proposes the following strategy: ; in, Represents An array of elements, each element contains the average value of the output data, according to the size of RMB, in the data set containing data points Conduct experiments in . Indicates parameters, Represents an element of a single array output of size 1 per QMB, where To optimize the efficiency of the path planning algorithm, the present invention considers three parameters, namely, path cost, number of search nodes in the grid graph, and time complexity. Next, to measure the importance of the parameters, each data matrix is normalized with equal weights using the above formula and classified into Get a range array Then, sum each index data point in all result data matrices and merge them according to the total number of arrays (3 in this invention) to obtain Finally, the best .

[0022] S5. Based on the above strategies, determine the optimal planning path.

[0023] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A robot path planning optimization method based on the improved A-STAR algorithm is characterized by: The following steps are involved: S1: Use the A-STAR path planning algorithm to initialize the target location node; S2: Check the position of adjacent grid nodes and determine the target node ; S3: Search using optimized RMB; S4: Using the adaptive cost function, select the node with the lowest total cost and update it; S5: Select the updated node with the lowest total cost and determine the optimal path planning.

2. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 1 is characterized in that: The step S1 includes the following steps: S11: First, initialize the starting node to a time node with a cost of 0; S12: Add the starting node to the OPEN list, which is used to store nodes to be expanded.

3. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 1 is characterized in that: In the step S2, it includes the following steps: S21: Search surrounding nodes on the weighted graph; S22: If the surrounding nodes are obstacle nodes , then ignore the node; S23: If the surrounding nodes are not obstacle nodes, add them to the OPEN list; S24: When the list is not empty, select the node with the lowest total cost and delete it from the OPEN list.

4. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 3 is characterized in that: In the step S24, the cost function is Calculate the node cost, where Represents a heuristic function that uses the Euclidean distance.

5. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 4 is characterized in that: The step S24 specifically includes the following steps: S241: Search and generate all possible successor nodes of the selected node, and calculate their costs and heuristic values according to the cost function; S243: If the selected node is the target node, stop searching and return the optimal path; S244: For each successor node, if it has not been visited or has a higher cost, update its cost and heuristic value; S245: If each updated successor node does not already appear in the list, add it to the list; S246: If the successor node is already in the list and the new value is lower than the previous value, update its cost and heuristic value; if the successor node is already in the visit list, ignore the node; S247: Repeat the above steps until every node on the searched weighted grid graph is traversed.

6. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 1 is characterized in that: In the S3 step, the robot uses a unit block (RMB) to search for adjacent nodes. The unit block consists of eight units surrounding the robot.

7. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 6 is characterized in that: In the step S3, the unit block is represented as ;in, Indicates the size of the motion block, Indicates the cost of movement in the four directions of up, down, left, and right; Represents the movement cost in the four diagonal directions; represents the cost associated with the four basic movements in the directions of up, down, left, and right, represents the costs associated with movement in the four diagonal directions of up-right, up-left, down-right, and down-left; represents the adaptive cost.

8. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 1 is characterized in that: In the step S4, the adaptive cost function is ,in, Represents the current node, represent The cost of moving represents the motion matrix of the node, represents the target node, Used to increase the importance of the distance value between each motion matrix node and the target node; in order to determine the motion matrix node and The cost of the , formulated an adaptive cost function , which multiplies the basic movement and diagonal Costs associated with exercise.

9. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 8, characterized in that: The S4 step specifically includes the following steps: S41: First, the cost of the current node With the current node and Motion Matrix nodes Add the Euclidean distance between them to get the absolute difference between the x-coordinate and the y-coordinate; S42: With the motion matrix Nodes and target nodes Add the Euclidean distances between them and multiply by a constant ; In the step S42, Used to increase the importance of the distance value between each node of the motion matrix and the target node.

10. The robot path planning optimization method based on the improved A-STAR algorithm according to claim 9 is characterized in that: In the step S4, in order to adapt to the different output results of the motion blocks caused by the different motion blocks of the robot, the method for obtaining the optimal motion block adopts the formula ,in, Represents An array of elements, each element contains the average value of the output data, according to the size of RMB, in the data set containing data points Conduct experiments in Indicates parameters, Represents an element of a single array output of size 1 per QMB, where ; The specific method is: A1: Use the formula to equally weight each data matrix and put it into the range of R=1000 to obtain a range array. ; A2: Sum each index data point in all result data matrices and merge them by the total number of arrays to obtain ; A3: Determine the best by selecting the minimum value of the final result data array for each RMB .

Images