Path Planning For Cooperative Searching Of Multi-micro Robots

The shortest path of robot cluster searching for multiple target is a hot research topic in path planning.This thesis aims to solve the shortest circuit path for the robot to search the targets,we use swarm intelligence algorithm to reduce computation time in solving large-scale traveling salesman problem,so as to improve the efficiency of the robot to complete the search task.Firstly,the clustering algorithm was used to cluster the targets,the points with similar clustering characteristics were grouped into groups,and the large-scale TSP problem was decomposed into a small-scale single TSP problem for solving.Compared with k-means clustering algorithm,Fuzzy c-means clustering algorithm(FCM)shows better performances in the subsequent shortest path solution.For each set of search points,two swarm intelligence algorithms,hybrid particle swarm optimization(PSO)and genetic algorithm(GA),were used to solve the problem,and the results and convergence speed of the algorithm were compared.Aiming at the convergence time problem of the two algorithms,the improved circle algorithm is used to first optimize the population individuals,and then the obtained individuals are solved by genetic algorithm,which significantly improves the convergence speed of the optimal solution and reduces computation time.Finally,We use tuttlebot3 with ROS to conduct the simulation route solution experiment,and the robots complete Search Tasks.Through the improvement and comparison of the two intelligent algorithms,the intelligent algorithm used in this paper has achieved good results in solving the total distance of the shortest circuit path of the robot and the longest path of the grouping problem.