The Research On Path Planning Technology Based On Automated Guided Vehicle System

With the rapid development of automation and AI in industry,the goal of smart factory is becoming more and more urgent.Automated guided vehicle(AGV),as an important transportation tool in smart factory,has been widely applied in smart factory.The AGV system is mainly applied in the material transportation link,which is convenient for the factory's automatic management,saving manpower cost and improving the production efficiency of the workshop.This paper is based on the development of AGV prototype and control management system based on the cooperative project,and focuses on the in-depth study of the path planning technology in combination with the theory and practice.The main contents are as follows:First of all,the development of AGV prototype system,the main research contents include: hardware selection,control system construction,program development and debugging,the development of the upper computer management system,and the AGV system structure includes two parts: upper computer and lower computer.The upper computer is responsible for the map model management,the path planning and the task management,the lower computer is the autonomous guide transport vehicle,including the guidance system,the motion control system,the walking mechanism,the safety device,the hydraulic system and so on.We plan the whole system execution process,and make instruction control and information transmission between host computer and slave computer through wireless communication.Secondly,the part of the core research content path planning is deeply studied.The path planning method is divided into two kinds of methods: global planning and local planning.The global path planning method is studied.Three modeling methods are compared,including topological graph method,visual graph method and grid graph method.The grid graph method is selected as the map modeling method in this paper,and the Dijkstra algorithm,the A* algorithm and the JPS calculation are compared.The advantages and disadvantages of the three algorithms,select the most efficient JPS algorithm,improve the heuristic function,jump search rule optimization and path trajectory optimization,and finally get the optimized global programming JPS algorithm,and study the local path planning method: by comparing the advantages and disadvantages of RRT algorithm,DWA algorithm and APF algorithm,we choose the local path planning DWA which is most suitable for the development of this project.The fuzzy inference control method is adopted for the combination of the evaluation coefficient combination(alpha,beta,gamma),and the evaluation coefficient combination is adjusted according to the environmental information in real time,so that the DWA algorithm can be maximized to the complex working environment.Finally,the simulation experiments of the global and local programming algorithms are carried out to verify the effectiveness of the algorithm.Finally,the global planning and local planning are integrated,and a complete path planning system is obtained.In any complex environment variable,the optimal path is planned through the fusion of the optimized JPS algorithm and the optimized DWA algorithm.Then the mobile robot TurtleBot is used as the experimental model,and the path planning of the JPS algorithm and the DWA algorithm is fused in the laboratory.The method is verified by practice,and achieves good practical results.The algorithm is effective,and it can perfectly solve most of the path planning problems.