Research On Task Scheduling And Path Planning Of Multi-AGV Warehouse System

With the rapid development of the logistics industry and the centralized cluster of warehousing distribution,the logistics and warehousing industry uses warehousing AGVs to replace labor on a large scale,and the number of AGVs used and warehouse tasks are increasing day by day,putting forward higher requirements for the warehousing system.Through the research and improvement of multi-warehouse AGV path planning and task allocation algorithm,the main goal is to minimize the total working time of multi-warehouse AGV,coordinate the conflicts among various AGVs,avoid collisions and deadlocks,and improve the work efficiency and throughput of warehouse system.The main research contents are as follows:(1)Use hybrid control structure to control each part,use grid method to build map model,set priority for tasks,and build task list model.Multi-sensor fusion is used to locate and navigate the robot,inertial sensor and vision sensor fusion is used to determine the position of the robot,and lidar is used to identify the unknown obstacles in real time.(2)The task cost function is constructed,and the objectives and constraints of task allocation are mathematically modeled according to the warehouse environment.A clustering weighting auction method is proposed to select the appropriate weighting coefficient by simulation comparison.To solve the problem of uneven task allocation caused by the large map,the initial task allocation was carried out before the task allocation began.(3)The improved A* algorithm is used to carry out global path planning.Compared with the original algorithm,the improved A* algorithm can reduce the number of path turns planned,avoid congested sections,and greatly reduce the long path finding time.In order to avoid unknown obstacles in the working path,DWA algorithm is used for local path planning,and common obstacle types are simplified and simulated.(4)Analyze the coordination problems of AGV path conflicts,classify the conflicts,study them based on the time window algorithm,and analyze the time window models corresponding to different conflict situations.Dynamic priority of AGVS is determined according to priority criteria.According to different conflict situations,conflict coordination strategies such as stop waiting,time window right shift and path replanning are adopted to plan deadlock-free and conflict-free multi-AGV paths.