Robot Multi-Objective Path Planning Method Based on Self-Learning Evolutionary Algorithm

Abstract

The present invention relates to the technical field of mobile robot path planning, in particular to a multi-objective path planning method for robots based on a self-learning evolutionary algorithm. It includes: initializing algorithm parameters; initializing the population; in the employed bee stage, performing path crossover, path mutation, path shortening, and path safety operations on each solution in the population; calculating the objective value of each solution, and dividing the population into a non-dominated solution set and a dominated solution set; in the onlooker bee stage, acting on the non-dominated solution set based on a collaborative learning mechanism and acting on the dominated solution set based on a domination-guided learning mechanism; merging the non-dominated solution set and the dominated solution set; in the scout bee stage, using an individual restart strategy to act on the solutions whose consecutive evolution failure times exceed the set maximum threshold; updating the non-dominated solution set and determining whether the termination condition is reached. The application of the present invention can effectively improve the path planning efficiency and enhance the path quality.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile robot path planning, and particularly to a multi-objective path planning method for a robot based on a self-learning evolutionary algorithm. Background Art

[0002] With the rapid development of science and technology, mobile robots have been widely used in many fields such as industrial manufacturing, medical surgery, agricultural production, and autonomous driving. Path planning is a key technology for realizing the autonomous navigation of mobile robots, and its core task is to search for an optimal or sub-optimal feasible path for the robot in the configuration space from the starting position to the target position. The path planning methods in the prior art usually only optimize for a single path length objective. However, when performing path planning in complex actual scenarios, multiple factors need to be comprehensively considered, especially potential dangerous obstacles in the environment. Therefore, the existing single-objective path planning methods are difficult to generate high-quality paths that meet the actual needs.

[0003] In addition, the existing methods for solving path planning lack specific optimization operations for problems and objectives, resulting in poor accuracy of the solutions generated by the methods. At the same time, most methods focus on the individual evolution of individuals in the population and ignore the co-evolution between individuals, resulting in the underutilization of excellent information, thereby affecting the quality of the final solution. Therefore, for the multi-objective path planning problem, there is an urgent need for a multi-objective path planning method based on a self-learning evolutionary algorithm to further improve the quality of the solution. Summary of the Invention

[0004] The object of the present invention is to provide a multi-objective path planning method for a robot based on a self-learning evolutionary algorithm, which solves the problems of minimizing the path length and maximizing the safety during the path planning process, so as to achieve the purpose of effectively improving the path planning efficiency and enhancing the path quality.

[0005] A multi-objective path planning method for a robot based on a self-learning evolutionary algorithm provided by the present invention is characterized by including the following steps:

[0006] Step 1, initialize the algorithm parameters, and set the population size Ps and the maximum number of consecutive mutation failures in the path mutation operation C mf and the maximum number of consecutive evolution failures C ef and the cut-off running time;

[0007] Step 2, initialize the population, and generate Ps initial solutions by using a hybrid initialization strategy;

[0008] Step 3, employed bee stage, sequentially perform path crossover operation and path mutation operation to explore each solution in the population, randomly select to perform path shortening operation or path safety operation to evolve the solution, and accept the evolved solution as the new solution;

[0009] Step 4, calculate the objective values of the solutions in the population, perform non-dominated sorting on the solutions in the population, and divide the population into non-dominated solution set and dominated solution set;

[0010] Step 6, follower bee stage, use two self-learning optimization mechanisms, including collaborative-based learning mechanism and domination-guided learning mechanism, to act on the non-dominated solution set and the dominated solution set respectively to achieve in-depth search of the solution space;

[0011] Step 9, merge the non-dominated solution set and the dominated solution set;

[0012] Step 7, scout bee stage, use the individual restart strategy to act on the solutions whose consecutive evolution failure times exceed the set maximum threshold to enhance the vitality of the population;

[0013] Step 8, update the non-dominated solution set, judge whether the termination condition is reached, if so, end the evolution and output the non-dominated solution set, otherwise return to Step 3.

[0014] Furthermore, in Step 2, the process of the hybrid initialization strategy is as follows: use the heuristic algorithm to generate the first solution, and then generate the remaining Ps - 1 solutions according to the population size using the rapidly-exploring random tree algorithm and the space partitioning method; one solution represents a path, and the path is represented by a discrete sequence, Path = P 0 ,... P i ,... P n+1 , Path represents a path, P 0 represents the first node in the sequence, i.e., the starting node, P n+1 represents the last node in the sequence, i.e., the target node, represents an intermediate node, i represents the index number of the intermediate node, and the value range is 1 ≤ i ≤ n , n represents the number of intermediate nodes in the path; each solution corresponds to two objective values, namely path length and path safety; the path length calculation formula is where represents the length of the path Path , represents the Euclidean distance between two consecutive points in the path; the calculation formula for path safety is , where represents the safety of the path, where is an obstacle in the configuration space, j represents the index number of the obstacle, m is the number of obstacles, is the minimum distance between the path and the obstacle in the configuration space.

[0015] Furthermore, in step 3, a solution in the population is used as the first input solution for the path crossover operation, and a solution is randomly selected from the population as the second input solution. The path crossover operation includes the following steps: find the intermediate nodes with the closest distance to each other in the discrete sequences corresponding to the two input solutions, and set the found nodes as in the first input solution and in the second input solution. Among them, q is the selected node index in the sequence corresponding to the second input solution. Check whether the straight line segment between and intersects with the obstacle. If it does not intersect, generate two new solutions: the first new solution is composed of the part from the starting node of the sequence corresponding to the first input solution to node , and the part from node to the target node in the sequence corresponding to the second input solution is spliced; the second new solution is composed of the part from the starting node of the sequence corresponding to the second input solution to node , and the part from node to the target node in the sequence corresponding to the first input solution is spliced. Perform non-dominated sorting on the two input solutions and the two new solutions, select the non-dominated solutions, and randomly select one solution from them for output; if it intersects, directly output the first input solution.

[0016] Furthermore, in step 3, the solution output by the path crossover operation is used as the input solution for the path mutation operation. The steps of the path mutation operation are as follows: set a counter to record the number of consecutive mutation failures, and set its initial value to 0. Randomly select an intermediate node in the sequence corresponding to the input solution, and determine the minimum distance between node d min and the surrounding obstacles. Inside the circle with node as the center and d min as the radius, randomly generate a new node P new ; check whether node P new intersects with node and node Whether the straight line segment between the nodes intersects with the obstacle. If not, the new node P new Replacement node constitutes a new solution. Calculate the objective value of the new solution. If the objective value of the new solution is less than the objective value of the input solution, output the new solution and the path mutation operation ends; if the objective value of the new solution is higher than the objective value of the input solution, increment the counter and generate a new node again; if it intersects, increment the counter and generate a new node again; when the counter reaches the maximum number of consecutive mutation failures C mf and no new solution with a smaller objective value is generated, output the input solution.

[0017] Furthermore, in step 3, the solution output by the path mutation operation is used as the input solution for the path shortening operation. The steps of the path shortening operation are as follows: set the input solution as the new solution; randomly select two non-adjacent intermediate nodes in the sequence corresponding to the new solution P x and P y , check whether the straight line segment between the nodes P x and P y intersects with the obstacle; if not, delete the nodes P x and P y between them, the path shortening operation ends and the new solution is output; if there is a collision, continue to randomly select non-adjacent intermediate nodes until all non-adjacent pairs of nodes are traversed and the new solution is output.

[0018] Furthermore, in step 3, the solution output by the path mutation operation is used as the input solution for the path safety operation. The steps of the path safety operation are as follows: in the sequence corresponding to the input solution, an adjacent pair of nodes forms a line segment. Find the line segment closest to the obstacle , and determine the position on the line segment closest to the obstacle as the node P nearest , determine P nearest the node on the surface of the obstacle closest to it P obstacle ; calculate the distance P nearest between the node P obstacle and the node d min , starting from the node P nearest as the starting point, in the direction of , at a distance from the nodeP nearest d min Generate a new node at the position of P new , check the node P new and the node and the node to check whether the straight line segment between them intersects with the obstacle; if not, insert the node P new between the node and the node to form a new solution and output the new solution; if it intersects, output the input solution.

[0019] Furthermore, in step 5, the collaborative learning mechanism acts on the non-dominated solution set, and the dominance-guided learning mechanism acts on the dominated solution set, where

[0020] The collaborative learning mechanism is to calculate the objective values of each solution in the non-dominated solution set, and normalize these objective values according to the objective type using the following formulas respectively,

[0021]

[0022] refers to the k th objective value of the l th solution. In addition, and represent the maximum and minimum values of the l th objective of all solutions respectively. Therefore, represents the result of normalizing the k th objective value of the l th solution; for each solution in the non-dominated solution set, perform the following steps: if the current solution obtains the minimum value in terms of path length or path safety, give priority to performing path shortening operation on it to generate a new solution; then apply path safety operation to the generated solution to further improve the safety of the solution; for the solutions that do not obtain the minimum value in terms of path length or path safety, perform optimization operation according to the size of the normalized value: if the normalized value of the path length is higher than the normalized value of the path safety, perform path shortening operation on the current solution; otherwise, perform path safety operation to obtain the optimized non-dominated solution; if the optimized non-dominated solution dominates the non-dominated solution before optimization, replace the non-dominated solution before optimization with the optimized non-dominated solution to generate and output the optimized non-dominated solution set;

[0023] The learning mechanism based on domination guidance is as follows: for each dominated solution, a non-dominated solution is randomly selected from the non-dominated solution set for comparison; if both the path length and path safety of the non-dominated solution are less than those of the current dominated solution, the path shortening operation and path safety operation are sequentially performed on the current dominated solution; if the path length of the non-dominated solution is less than that of the current dominated solution, the path shortening operation is performed on the current dominated solution; otherwise, the path safety operation is performed on the current dominated solution to generate an optimized dominated solution. If the path length and path safety of the optimized dominated solution are less than those of the dominated solution before optimization, the optimized dominated solution is used to replace the dominated solution before optimization, and the optimized dominated solution set is generated and output.

[0024] Furthermore, in step 7, the individual restart strategy is as follows: non-dominated sorting is performed on the population generated in the follower bee stage, and the population is divided into a non-dominated solution set and a dominated solution set; for each solution in the dominated solution set, it is detected whether the number of consecutive evolutionary failures exceeds the set maximum threshold. If it is satisfied, the current dominated solution is replaced with a new solution, and the new solution is randomly generated in any of the following ways. One way is to select two solutions from the non-dominated solution set, one of which is the non-dominated solution with the minimum value in terms of path length or path safety, and the other is a random non-dominated solution, and the path crossover operation is performed on these two solutions to generate a new solution; the other way is to use the fast random tree algorithm to generate five new solutions, and the solution with the shortest path length is selected as the new solution.

[0025] The multi-objective path planning method for robots based on the self-learning evolutionary algorithm provided by the present invention simultaneously optimizes two objectives, namely path length and path safety, for the dangerous obstacles existing in the path planning process. Secondly, the present invention adopts a hybrid initialization strategy to generate a high-quality initial population. Compared with some existing algorithms, the present invention proposes optimization operations for two objectives, improves the exploration ability of the algorithm for the solution space, and at the same time proposes two self-learning optimization mechanisms to guide the in-depth evolution of solutions. In addition, the individual restart strategy used by the present invention enhances the vitality of the population. In summary, the present invention can effectively improve the quality of solutions and greatly improve the efficiency of path planning. Brief Description of the Drawings

[0026] Figure 1 is the implementation flowchart of the present invention;

[0027] Figure 2 The indoor home scene used in the test of the present invention Figure 1 is the schematic diagram of the features;

[0028] Figure 3 The wide-area street scene used in the test of the present invention Figure 2 is the schematic diagram of the features;

[0029] Figure 4 For the solution distribution diagram obtained by comparing the present invention with the existing algorithms in the ground Figure 1 ;

[0030] Figure 5 For the solution distribution diagram obtained by comparing the present invention with the existing algorithms in the ground Figure 2 ;

[0031] Figure 6 For the convergence curve diagram of the present invention compared with the existing algorithms on the ground Figure 1 ;

[0032] Figure 7 For the convergence curve diagram of the present invention compared with the existing algorithms on the ground Figure 2 ;

[0033] Figure 8 For the solution display diagram with the best path length and path safety obtained by the present invention on the ground Figure 1 ;

[0034] Figure 9 For the solution display diagram with the best path length and path safety obtained by the present invention on the ground Figure 2 ; Specific embodiments

[0035] As Figure 1 shown, the multi-objective path planning method for a robot based on a self-learning evolutionary algorithm provided by the present invention is mainly implemented through the following steps.

[0036] Step 1, initialize the algorithm parameters. Set the population size Ps , the maximum number of consecutive mutation failures in the path mutation operation C mf , the maximum number of consecutive evolution failures C ef , where Ps = 50, C ef = 20, C mf = 10, and the cut-off running time is 50s, which is the termination condition of the algorithm.

[0037] Step 2, initialize the population. Generate Ps initial solutions by using a hybrid initialization strategy. Specifically, a solution represents a path, and the path is represented by a discrete sequence, Path = P 0 , ..., P i , ..., P n+1 , Path represents a path, P 0Represents the first node in the sequence, i.e., the starting node, P n+1 Represents the last node in the sequence, i.e., the target node, Represents an intermediate node, i Represents the index number of the intermediate node, and the value range is 1 ≤ i ≤ n , n Represents the number of intermediate nodes in the path; each solution corresponds to two objective values, namely the path length and the path safety; the path length calculation formula is , where Represents the path Path length, Represents the Euclidean distance between two consecutive points in the path; the path safety calculation formula is , where, Represents the safety of the path, where is an obstacle in the configuration space, j Represents the index number of the obstacle, m is the number of obstacles, is the minimum distance between the path and the obstacles in the configuration space. The hybrid initialization strategy steps are as follows. Using an efficient heuristic algorithm, the A* algorithm is used in the present invention to generate the first initial solution, and then the Rapidly-exploring Random Tree (RRT) algorithm and the space partitioning method are used to generate the remaining Ps −1 initial solutions. Specifically, two candidate solutions are generated respectively using the RRT algorithm and the space partitioning method, and the non-dominated solution is selected as the next initial solution, finally generating diverse initial solutions, providing a good basis for subsequent optimization.

[0038] Step 3, employed bee stage, successively perform path crossover operation and path mutation operation, explore each solution in the population, randomly select to perform path shortening operation or path safety operation to evolve the solution, and accept the evolved solution as the new solution. The random selection method in the technical solution of the present invention is: by using the random function rand () to generate a random number, and make different selections according to the value of the random number. For example, in Step 3, by using the random function rand () to generate a random number between 0 and 1. If the random number is greater than 0.5, perform the path shortening operation on the solution, otherwise, perform the path safety operation on the solution.

[0039] Specifically, take the current solution in the population as the first input solution for the path crossover operation, randomly select a solution in the population as the second input solution. The path crossover operation includes the following steps: find the intermediate nodes with the closest distance to each other in the discrete sequences corresponding to the two input solutions, and set the found node as in the first input solution and , where q is the selected node index in the sequence corresponding to the second input solution. Check whether the straight line segment between and intersects with the obstacle. If not, generate two new solutions: The first new solution is composed of the part from the starting node of the sequence corresponding to the first input solution to node , concatenated with the part from node to the target node in the sequence corresponding to the second input solution; The second new solution is composed of the part from the starting node of the sequence corresponding to the second input solution to node , concatenated with the part from node to the target node in the sequence corresponding to the first input solution. Perform non-dominated sorting on the two input solutions and the two new solutions, select the non-dominated solutions, and randomly select one solution from them for output; If it intersects, directly output the first input solution.

[0040] Specifically, the solution output by the path crossover operation is used as the input solution for the path mutation operation. The steps of the path mutation operation are as follows: Set a counter to record the number of consecutive mutation failures, and set its initial value to 0. Randomly select an intermediate node in the sequence corresponding to the input solution, and determine the minimum distance between node d min and the surrounding obstacles. Generate a new node randomly within a circle with node d min as the center and P new as the radius; Check whether the straight line segment between node P new and nodes and intersects with the obstacle. If not, the new node P new replaces node to form a new solution. Calculate the objective value of the new solution. If the objective value of the new solution is less than the objective value of the input solution, output the new solution and end the path mutation operation; If the objective value of the new solution is higher than the objective value of the input solution, increment the counter and generate a new node again; If it intersects, increment the counter and generate a new node again; When the counter reaches the maximum number of consecutive mutation failures C mf and no new solution with a smaller objective value is generated, output the input solution.

[0041] Specifically, the solution output by the path mutation operation is used as the input solution for the path shortening operation. The steps of the path shortening operation are as follows: Set the input solution as the new solution; Randomly select two non-adjacent intermediate nodes in the sequence corresponding to the new solutionP x and P y , check whether the straight-line segment between the nodes P x and P y intersects with the obstacle; if not, delete the node P x and P y between, the path shortening operation ends, and output the new solution; if there is a collision, continue to randomly select non-adjacent intermediate nodes until all non-adjacent pairs of nodes are traversed, and output the new solution.

[0042] Specifically, the solution output by the path mutation operation is used as the input solution for the path safety operation. The steps of the path safety operation are as follows: in the sequence corresponding to the input solution, an adjacent pair of nodes forms a line segment, and find the line segment that is closest to the obstacle , and determine the line segment Set the position closest to the obstacle of the line segment P nearest as the node P nearest Determine the node on the surface of the obstacle closest to P obstacle ; calculate the distance between the node P nearest and the node P obstacle d min , starting from the node P nearest as the starting point, in the direction of , at a distance from the node P nearest d min generate a new node P new , check whether the straight-line segment between the node P new and the node and the node intersects with the obstacle; if not, insert the node P new between the node and the node to form a new solution, and output the new solution; if it intersects, output the input solution.

[0043] Step 4, calculate the objective value of the solutions in the population, perform non-dominated sorting on the solutions in the population, and divide the population into non-dominated solution sets and dominated solution sets.

[0044] ​Step 5, Follow-up Bee Phase, using two self-learning optimization mechanisms, including a cooperation-based learning mechanism and a domination-guided learning mechanism. The cooperation-based learning mechanism acts on the non-dominated solution set, and the domination-guided learning mechanism acts on the dominated solution set. Among them,

[0045] The cooperation-based learning mechanism is as follows: Calculate the objective values of each solution in the non-dominated solution set, and normalize these objective values according to the objective type using the following formulas respectively.

[0046]

[0047] refers to the k th solution's l th objective value. In addition, and represent the maximum and minimum values of the l th objective of all solutions respectively. Therefore, represents the result of normalizing the k th solution's l th objective value; For each solution in the non-dominated solution set, perform the following steps: If the current solution achieves the minimum value in terms of path length or path safety, then preferentially perform path shortening operation on it to generate a new solution; Subsequently, apply path safety operation to the generated solution to further improve the safety of the solution; For solutions that do not achieve the minimum value in terms of path length or path safety, perform optimization operations according to the size of the normalized value: If the normalized value of the path length is higher than the normalized value of the path safety, then perform path shortening operation on the current solution; Otherwise, perform path safety operation to obtain an optimized non-dominated solution; If the optimized non-dominated solution dominates the non-dominated solution before optimization, then replace the non-dominated solution before optimization with the optimized non-dominated solution to generate and output the optimized non-dominated solution set.

[0048] The domination-guided learning mechanism is as follows: For each dominated solution, randomly select a non-dominated solution from the non-dominated solution set for comparison; If the path length and path safety of the non-dominated solution are both less than those of the current dominated solution, then perform path shortening operation and path safety operation on the current dominated solution in sequence; If the path length of the non-dominated solution is less than that of the current dominated solution, then perform path shortening operation on the current dominated solution; Otherwise, perform path safety operation on the current dominated solution to generate an optimized dominated solution. If the path length and path safety of the optimized dominated solution are less than those of the dominated solution before optimization, then replace the dominated solution before optimization with the optimized dominated solution to generate and output the optimized dominated solution set.

[0049] Step 6, Combine the non-dominated solution set and the dominated solution set.

[0050] Step 7: Scout bee stage. The individual restart strategy is applied to the solutions whose consecutive evolution failure times exceed the set maximum threshold to enhance the population vitality. Specifically, the individual restart strategy is as follows: perform non-dominated sorting on the population generated in the onlooker bee stage, and divide the population into a non-dominated solution set and a dominated solution set; for each solution in the dominated solution set, detect whether the number of consecutive evolution failures exceeds the set maximum threshold. If it is satisfied, replace the current dominated solution with a new solution, and randomly select any one of the following methods to generate a new solution. One is to select two solutions from the non-dominated solution set, one of which is the non-dominated solution with the minimum path length or path safety, and the other is a random non-dominated solution, and perform path crossover operation on these two solutions to generate a new solution; the other is to use the fast random tree algorithm to generate five new solutions, and select the solution with the shortest path length as the new solution.

[0051] Step 8: Update the non-dominated solution set and determine whether the running time reaches 50 s. If it reaches, end the evolution and output the non-dominated solution set; otherwise, return to Step 3.

[0052] A multi-objective path planning method for robots based on a self-learning evolutionary algorithm proposed by the present invention combines two self-learning mechanisms with the artificial bee colony algorithm framework, named EABC. It improves the quality of the initial population through a hybrid initialization strategy, improves the global exploration ability of the algorithm through path crossover, path mutation, path shortening and path safety operations in the employed bee stage, improves the local search ability of the algorithm through two self-learning optimization mechanisms in the onlooker bee stage to achieve in-depth evolution of the solution, and enhances the population vitality through a new individual restart strategy in the scout bee stage.

[0053] Next, the present invention will be further described and illustrated by listing two specific embodiments of the present invention.

[0054] As Figure 2 and Figure 3 shown, Figure 2 is a home indoor scene, Figure 3 is a wide-area street scene. The sizes of the two scene maps are both standardized to 500×500 pixels, which are used to test and verify the effectiveness of the path planning method of the present invention. The present invention is experimentally compared and analyzed with the improved artificial bee colony algorithm (IMO-ABC), the improved multi-objective optimization algorithm (LMOEA / D), the non-dominated sorting algorithm (NSGA-II), the improved multi-objective artificial bee colony algorithm (IMOABC) and the enhanced improved multi-objective particle swarm optimization algorithm (FIMOSO). The simulation experiments and results are as follows: In the two scenes, the present invention and the above five existing algorithms are respectively run for 50 s, and the non-dominated solutions obtained are as Figure 4 and Figure 5 shown. In Figure 4 and Figure 5In the figure, the horizontal axis represents the safety index value, and the vertical axis represents the length index value. It can be seen that the non-dominated solution set obtained by this method is closer to the lower left corner of the picture, indicating that the application of the present invention can obtain a non-dominated solution set of higher quality.

[0055] As Figure 6 and Figure 7 shown, the comparison results of the present invention and the above five existing algorithms in terms of hypervolume metric. Hypervolume represents the volume of the hypercube formed by the space between the non-dominated solution set and the reference point, where the reference point is the point corresponding to the worst values obtained in terms of path length and path safety during the algorithm operation. Hypervolume is used to evaluate the comprehensive performance of the algorithm, and its value range is [0,1]. The larger the hypervolume value, the better the comprehensive performance of the algorithm. In Figure 6 and Figure 7 the figure, the horizontal axis represents the hypervolume value, and the vertical axis represents the running time. The algorithm of the present invention reaches the maximum hypervolume value in two test scenarios, which can reflect that the comprehensive performance of the method of the present invention is the best.

[0056] As Figure 8 and Figure 9 shown, the present invention generates the best paths for different optimization objectives in two test scenarios respectively. The figure intuitively shows the specific influence of the shortest path length and the highest path safety on the path planning result, thus revealing the differences in path selection for different optimization objectives. Therefore, the path planning method proposed by the present invention can achieve a better balance between path length and path safety and generate high-quality solutions.

Claims

1. A robot multi-objective path planning method based on a self-learning evolutionary algorithm, characterized in that: The following steps are included: Step 1: Initialize algorithm parameters and set population size Ps , the maximum number of consecutive mutation failures in a path mutation operation C mf , the maximum number of consecutive evolution failures C ef , the end of operation time; Step 2: Initialize the population and use a mixed initialization strategy to generate Ps The hybrid initialization strategy uses a heuristic algorithm to generate the first solution, and then uses a fast exploration random tree algorithm and a spatial partitioning method to generate the remaining solutions according to the population size. Ps -1 solution; a solution represents a path, and the path is represented by a discrete sequence, Path = [ P 0,..., P i , ..., P n+1 ], Path Represents a path, P 0 means the first node in the sequence, i.e. the starting node. P n+1 Indicates the last node in the sequence, i.e. the target node. represents an intermediate node, i Indicates the index number of the intermediate node, the value range is 1≤ i ≤ n , n Represents the number of intermediate nodes in the path; each solution corresponds to two target values, namely path length and path safety; the formula for calculating path length is: ,in Indicates the path Path Length, represents the Euclidean distance between two consecutive points in the path; the calculation formula for path security is ,in, represents the safety of the path, where is an obstacle in the configuration space, j Indicates the index number of the obstacle. m is the number of obstacles, is the minimum distance between a path and an obstacle in the configuration space; Step 3, the hired bee stage, performs path crossing operations and path mutation operations in sequence, explores each solution in the population, randomly selects to perform path shortening operations or path safety operations to evolve the solution, and accepts the evolved solution as the new solution; A solution in the population is used as the first input solution of the path crossover operation, and a solution is randomly selected in the population as the second input solution. The path crossover operation steps are to find the intermediate node with the shortest distance between the two input solutions in the discrete sequence corresponding to each other, and set the found node as the intermediate node in the first input solution. The second input solution ,in, q Check the selected node index in the sequence corresponding to the second input solution. and Whether the straight line segment between intersects with the obstacle, if not, two new solutions are generated: the first new solution is from the starting node of the sequence corresponding to the first input solution to the node , concatenate the sequence corresponding to the second input solution from node to the target node; the second new solution consists of the starting node to the node of the sequence corresponding to the second input solution , concatenate the sequence corresponding to the first input solution from node To the target node, perform non-dominated sorting on the two input solutions and the two new solutions, select the non-dominated solutions, and randomly select one solution from them to output; if they intersect, directly output the first input solution; The solution output by the path crossover operation is used as the input solution of the path mutation operation. The steps of the path mutation operation are to set a counter to record the number of consecutive mutation failures and set its initial value to 0. Randomly select an intermediate node in the sequence corresponding to the input solution. , and determine the node Minimum distance to surrounding obstacles d min , in the node is the center of the circle, d min Generate a new node randomly within a circle of radius P new ; Check node P new With Node and nodes Whether the straight line segment between nodes intersects with obstacles, if not, the new node P new Alternative Node A new solution is constructed, and the target value of the new solution is calculated. If the target value of the new solution is less than the target value of the input solution, the new solution is output and the path mutation operation ends. If the target value of the new solution is higher than the target value of the input solution, the counter is incremented by one and a new node is generated. If they intersect, the counter is incremented by one and a new node is generated. When the counter reaches the maximum number of consecutive mutation failures, C mf If no new solution with a smaller target value is generated, the input solution is output; The solution output by the path mutation operation is used as the input solution of the path shortening operation. The steps of the path shortening operation are as follows: set the input solution to the new solution; randomly select two non-adjacent intermediate nodes in the sequence corresponding to the new solution. P x and P y , check the node P x and P y Whether the straight line segment between the nodes intersects with the obstacle; if not, delete the node P x and P y The path shortening operation ends when the node between them is found, and a new solution is output; if there is a collision, non-adjacent intermediate nodes are randomly selected until all two non-adjacent nodes are traversed and a new solution is output; The solution output by the path mutation operation is used as the input solution of the path safety operation. The steps of the path safety operation are: in the sequence corresponding to the input solution, two adjacent nodes form a line segment, and find the line segment closest to the obstacle. , and determine the line segment The position closest to the obstacle is set as the node P nearest ,Sure P nearest The node on the surface closest to the obstacle P obstacle ; Compute nodes P nearest With Node P obstacle The distance between d min , in the node P nearest As a starting point, is the direction, distance from the node P nearest d min Create a new node at the location P new , check the node P new With Node and nodes Whether the straight line segment between them intersects with the obstacle; if not, the node P new Insert into node and nodes Between them, a new solution is formed and output; if they intersect, the input solution is output; Step 4, calculate the target value of the solution of the population, perform non-dominated sorting on the solutions in the population, and divide the population solutions into a non-dominated solution set and a dominated solution set; Step 5, the bee-following stage, uses two self-learning optimization mechanisms, including a collaborative learning mechanism and a dominant guidance learning mechanism, which act on the non-dominated solution set and the dominated solution set respectively to achieve a deep search of the solution space; Step 6, merge the non-dominated solution set and the dominated solution set; Step 7, the scout bee stage, uses the individual restart strategy to act on the solution whose number of consecutive evolution failures exceeds the set maximum threshold to enhance the vitality of the population; Step 8, update the non-dominated solution set and determine whether the termination condition is met. If so, end the evolution and output the non-dominated solution set. Otherwise, return to step 3.

2. A robot multi-objective path planning method based on a self-learning evolutionary algorithm according to claim 1, characterized in that: The learning mechanism based on collaboration is applied to the non-dominated solution set, and the learning mechanism based on dominance guidance is applied to the dominated solution set, where: The collaborative learning mechanism is to calculate the target value of each solution in the non-dominated solution set, and normalize these target values ​​using the following formula according to the target type: It refers to The solution target value, in addition, and Represents all solutions The maximum and minimum values ​​of the targets, therefore, Indicates The solution The result of normalizing the target values; perform the following steps on each solution in the non-dominated solution set: if the current solution obtains the minimum value in path length or path safety, the path shortening operation is preferentially performed on it to generate a new solution; then the path safety operation is applied to the generated solution to further improve the safety of the solution; for the solution that does not obtain the minimum value in path length or path safety, the optimization operation is performed according to the size of the normalized value: if the normalized value of the path length is higher than the normalized value of the path safety, the path shortening operation is performed on the current solution; otherwise, the path safety operation is performed to obtain the optimized non-dominated solution; if the optimized non-dominated solution dominates the non-dominated solution before optimization, the non-dominated solution before optimization is replaced by the optimized non-dominated solution, and the optimized non-dominated solution set is generated and output; The learning mechanism based on domination guidance is that for each dominated solution, a non-dominated solution is randomly selected from the non-dominated solution set for comparison; if the path length and path safety of the non-dominated solution are both smaller than the path length and path safety of the current dominated solution, the path shortening operation and the path safety operation are performed on the current dominated solution in sequence; if the path length of the non-dominated solution is smaller than the path length of the current dominated solution, the path shortening operation is performed on the current dominated solution; otherwise, the path safety operation is performed on the current dominated solution to generate an optimized dominated solution. If the path length and path safety of the optimized dominated solution are smaller than the dominated solution before optimization, the optimized dominated solution is used to replace the dominated solution before optimization, and the optimized dominated solution set is generated and output.

3. A robot multi-objective path planning method based on self-learning evolutionary algorithm according to claim 2, characterized in that: The individual restart strategy is to perform non-dominated sorting on the population generated in the follower bee stage, and divide the population into a non-dominated solution set and a dominated solution set; for each solution in the dominated solution set, check whether the number of consecutive evolution failures exceeds the set maximum threshold. If it meets the threshold, replace the current dominated solution with a new solution, and randomly select any of the following methods to generate a new solution: one is to select two solutions from the non-dominated solution set, one of which is a non-dominated solution that achieves the minimum value in path length or path safety, and the other is a random non-dominated solution. Path intersection operations are performed on these two solutions to generate new solutions; the other is to use the fast random tree algorithm to generate five new solutions, and select the solution with the shortest path length as the new solution.

Images