Font Size: a A A

Research And Application Of Path Planning For AGV Based On Particle Swarm Optimization And Artificial Potential Field Method

Posted on:2019-02-14Degree:MasterType:Thesis
Country:ChinaCandidate:Z D LiuFull Text:PDF
GTID:2428330566961552Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
Path planning is a key research area in AGV.For different environmental information,there are two research directions of path planning,that is,global static path planning known to environmental information and dynamic path planning in mobile environment.In view of the two path planning problems,this paper proposes the obstacle avoidance Particle Swarm Optimization and the global guidance artificial potential field respectively.Main contents of this thesis are as follows:1)Path planing for mobile robot by Particle Swarm Optimization is flaw with long path length,high arc cost under grid environment.an obstacle avoidance particle swarm optimization is proposed to solve these problems.Using the obstacle decomposition map method proposed in this paper,the environment is modeled,and based on the environmental model,the obstacle avoidance particle swarm optimization algorithm is used for path planning.The optimal N solutions of the fitness value obtained by the algorithm are used as alternative paths.Then the optimal particle dijkstra algorithm proposed by this paper performs accurate path search on these alternative paths,further improving path smoothness and reducing the path length..2)Aiming at the problem that the traditional dynamic path planning does not consider the path comprehensive performance.Considering the possible obstacles in robot workspace,this paper introduces the relative velocity between robot and obstacle into the potential field function,so as to solve the problem of dynamic obstacle avoidance.In addition,a known global path is introduced.The path node in the global path is used as the local target point of the robot,that is,the global path is used as a global guidance.Due to the existence of "global guidance",local mini-traps are avoided,and the information of the planned global optimal path can be fully utilized to ensure that the planned path performance is optimal,and at the same time,dynamic obstacle avoidance can be achieved.3)The AGV application system software is designed and developed according to the application requirements.The communication module,task management module,map management module,monitoring module,path planning module,and database module required by the upper computer are implemented.The obstacle avoidance particle swarm algorithm verifies the effectiveness of the algorithm and the utility of the system software.
Keywords/Search Tags:Path planning, AGV, Particle Swarm Optimization(PSO), Dijkstra algorithm, Artificial potential field
PDF Full Text Request
Related items