Research On Multi-AGV Autonomous Cooperative Scheduling Technology For Flexible Job Shop

With the development of digital workshop and unmanned factory technology,flexible manufacturing systems have become an important direction for the development of modern workshops.Automatic guided vehicle(AGV)has been greatly favored in the transformation of labor-intensive flexible automated production lines because it can reduce production costs,improve production efficiency and increase flexibility.The combined AGV combined with “mobile platform + robotic arm” can realize material storage,handling,and loading and unloading tasks for flexible operations.However,considering AGV automatic material replacement,automatic charging,and multi-AGV co-operation in the same area and other autonomous requirements,multi-AGV autonomous cooperative scheduling currently faces two urgent technical difficulties: task allocation and path planning.Regarding the issue above,this paper proposes a multi-AGV autonomous cooperative scheduling technology for flexible job shops,and develops a multi-AGV scheduling system.The main research contents are as follows:Analyzes the flexible job shop scenario and multi-AGV scheduling requirements are analyzed,and the multi-AGV autonomous collaborative scheduling problem is defined.The evaluation indicators of the scheduling is analyzed.The overall framework of multi-AGV autonomous collaborative scheduling is proposed by combining task allocation and path planning.In terms of autonomous collaborative scheduling task allocation,a multi-criteria evaluation scheme performed by TOPSIS and AHP is constructed;Further,a dynamic multiattribute weighted dispatch rule based on the AGV state domain,CNC state domain and task state domain are proposed for dynamic scenarios.The idea of dynamic weight adjustment is used to adapt the dispatch rule to different flexible operation scenarios.In addition,a static task allocation algorithm based on genetic and tabu is proposed for static scenarios.The tabu list is used to avoid circuitous search,and to ensure that the diverse and effective search is globally optimized.The simulation is performed,and analysis shows that the dynamic multi-attribute weighted dispatch rule has better scheduling indicators in a dynamic environment with multiple environmental parameters;at the same time,the improved genetic tabu algorithm has better scheduling indicators than the dispatch rule in a static environment.To improve genetic tabu algorithm scheduling capabilities in a dynamic environment,the task window strategy is used.In terms of autonomous collaborative scheduling path planning,aiming at the collisionfree path planning of multiple AGVs in the space and time dimensions,an offline collisionfree path planning method based on the time step network is used.The A* algorithm is used to search for collision-free paths in the network;Furthermore,for the changing scene of the AGV online environment,an online traffic management strategy based on pre-application and adaptive speed change is proposed;Aiming at the problems of encounters and head-on collisions in the online phase,an online conflict detection and re-planning strategy based on discrete time windows is proposed to further ensure the dynamic planning ability when conflicts occur.The path planning algorithm has been simulated and the results show that the time-step network-based collision-free path planning algorithm can effectively resolve multi-AGV encounter conflicts and head-on collision conflicts,and can plan paths without collisions in the space and time dimensions.For the development of a flexible AGV multi-AGV scheduling system,a system solution and system architecture are designed,and the key modules including a task scheduling module,a path planning module,and an application scenario module are described.The experimental platform was used to test the function and performance of the system.The results verified the effectiveness and superiority of the proposed autonomous cooperative scheduling technology.At the same time,the application results of the real workshop environment show that the proposed technology can meet the functional requirements of flexible job shop scenarios,and can achieve multi-AGV autonomous operation.