Research On Multi-AGV Dispatching System For Industrial Logistics

Industrial logistics refers to the process of transportation and turnover of tools,raw materials and finished products in the manufacturing process of products.AGV is applied to industrial logistics and can complete the transportation task according to the production rhythm of products.It can improve the transportation efficiency,reduce the production cost and further improve the production efficiency.It is an important link in the automation and intelligentization of enterprise production.The practical application of AGV in industrial logistics is often that multiple trolleys coordinate and orderly to carry out each transportation task,and accurately transport the goods to each work station.The scheduling control of each AGV in the execution process is completed by the background scheduling system.The background scheduling system needs to carry out motion control and state monitoring on the vehicle,so as to realize vehicle positioning,path planning,navigation and task scheduling,which is of great significance to the improvement of the transportation efficiency of the whole AGV system.In this paper,aiming at the problems that multiple AGVs need to be solved in the transportation of industrial logistics,refer to the relevant literature to understand the relevant research progress at home and abroad,and refer to the design methods of the previous researchers on the multiple AGV dispatching system,and then develop a set of multiple AGV dispatching system.The positioning algorithm,path planning algorithm,and scheduling algorithm have been studied,and the WEB application system of multiple AGV scheduling systems has been realized.The research content is as follows:(1)Research on the positioning algorithm of AGV,this paper use particle filter positioning algorithm to realize AGV global positioning,and design the motion state model and measurement observation model of AGV,and conduct experiments on the accuracy of multiple AGV positioning under simulation conditions.(2)Research and improve the AGV global path planning algorithm,improve the traditional A* algorithm,and add the time dimension to the path conflict problem that may occur in the multi-AGV scheduling process,and propose an improved A*algorithm based on the schedule.The test results show that the algorithm can effectively avoid all kinds of path conflicts.(3)For the multi-AGV scheduling problem in industrial logistics,this paper takes the shortest overall AGV transportation time and transportation balance in the processing process as the goal,and proposes a hybrid genetic algorithm based on timetable and A* algorithm.Model scheduling tasks to achieve optimal allocation of workshop tasks and AGVs,and effectively improve scheduling efficiency.(4)Based on the above-mentioned supporting algorithms,applying database and Web software development technology,a set of multiple AGV scheduling system was developed under the environment of ubuntu + ROS + WEB.The system realizes data resource storage and management,multiple AGV path planning,and task scheduling.,AGV status monitoring and other functions.In this paper,simulation tests and real environment experiments are carried out for multiple AGV scheduling systems.The AGV achieves autonomous positioning by relying on data collected by its own sensors,performs transportation tasks and completes scheduling tasks according to tasks distributed by the scheduling system and path planning scheme.The position of the test car and the working results are in line with the expected design objectives.