Research On Path Planning Algorithm Of Mobile Robot

With the development of computer perception technology and SLAM technology,the performance of mobile robots has been gradually improved and has been widely used in agricultural and industrial fields.Path planning technology is one of the important components of mobile robot system.Due to the complexity of the actual working environment and the robot motion model,traditional path planning methods have problems that are difficult to solve effectively.According to specific performance indicators,the robot finds the optimal collisionfree path from the starting position to the target position.Many route planning algorithms can plan paths well when information about the environment is already known.But in unknown environments,especially in complex environments wi th various irregular obstacles,many algorithms fail.The more complex the state information of the mobile robot is,the higher the requirements for path planning are correspondingly.Therefore,how to efficiently complete the path planning has become an urgent problem to be solved.Firstly,this thesis expounds the research background and significance of the subject,and at the same time explains the research status of mobile robot technology and path planning algorithms,and classifies the types of planning algorithms based on the state that the path planning algorithm is fully known or partially known based on the map environment information..Secondly,based on the principle of grid map construction method,this thesis constructs the simulation environment map of the algorithm,and establishes a mathematical model for the path planning problem of mobile robots.This thesis introduces and simulates in detail the Dijkstra algorithm,Genetic algorithm,Ant Colony Optimization,Probabilistic Roadmap Method,Rapidly-Exploring Random Tree algorithm in the global path planning algorithm of mobile robots,and Dynamic Window Approach,fuzzy algorithm in the local path planning algorithm.Finally,based on the typical algorithms in global path planning and local pa th planning,A~* algorithm and Artificial Potential Field algorithm,a combined planning algorithm is proposed.The algorithm combines the advantages of heuristic search and non-oscillating state of A~* algorithm with the advantages of high real-time performance,small computational load and strong path smoothness of Artificial Potential Field algorithm.Using MATLAB and V-REP simulation software,the mobile robot and the working environment map are modeled,and the correctness and rationality of the combined planning algorithm are simulated and verified.The simulation results show that the combined planning algorithm proposed in this thesis solves the problem that the Artificial Potential Field algorithm is prone to generate local minima in the face of clustered obstacles and the problem of low planning efficiency of the A~* algorithm.As a result,the robot can effectively escape the oscillating area and complete the planned tasks.