Research On Scheduling Method Of Multi AGV System In Unmanned Factory

With the rapid development of logistics information and automation technology,AGV automatic transportation system used for handling materials has been widely used in unmanned factory.AGV scheduling takes the specific system parameters as the optimization objective,carries on the weighted arrangement to the transportation task and the vehicle,and determines the reasonable allocation plan to realize the best matching between the transportation task and the vehicle.Whether the scheduling of the AGV system is reasonable or not will directly affect the cost and efficiency of the whole production system.In this paper,the following research work has been carried out for the scheduling problem of the AGV system in the unmanned factory:(1)Taking DrRobot i90 wireless intelligent robot as the object of this study,the AGV mathematical model driven by two wheel differential drive is established.And the model of AGV navigation system based on the dead reckoning is set up to realize the real-time positioning of the AGV in the system.Then the position and orientation information of the AGV in the coordinate system of its operating environment is determined.(2)According to the functional requirements of AGV system,the software architecture of AGV scheduling system based on the distributed control principle is designed.The effects of user layer,function layer,support layer and data layer are expounded,and the implementation method of management function,communication function,task scheduling function and AGV scheduling function are elaborated respectively.The working process of AGV scheduling system based on four levels and four functions is analyzed.(3)According to the actual production environment of the unmanned factory with automated warehouse and automated workshop,the single AGV path planning problem from the delivery of workpieces to the completion of the process is studied.In the automatic warehouse,the traveling salesman problem based on genetic algorithm is used to solve the shortest path of AGV;in the automation workshop,taking into account the processing path requirement of workpiece processing,the improved genetic algorithm is used to solve the shortest path of AGV.Then,the improved artificial potential field method is adopted for the path planning of AGV,which can make AGV reach the target point successfully.(4)The multi AGV system scheduling problem of unmanned factory is studied.With the total travel time of AGVs as the optimization goal,the problem of traffic conflicts of multi AGV system in the running process is analysised,and the coordinated scheduling strategies are put forward to solve traffic problems.The scheduling problem of multi AGV system is solved by modified genetic algorithm.(5)The AGV scheduling system developed in this paper is verify.In order to realize the remote control and monitoring of AGV system,the human-computer interaction system software of AGV scheduling system is developed,and the simulation and experiment of single AGV system and multi AGV system is carried out respectively.The simulation and experimental results show that the AGV scheduling system designed in this paper is correct and feasible,and can successfully complete the tasks assigned by the host computer system.