Application And Optimization Of Scheduling Algorithms In Storage Intelligent AGV

With the rapid development of the intelligent manufacturing industry,the requirements of its dependence on the intelligent storage industry are increasing day by day,and the automatic guided vehicle(Automated-Guided-Vehicle,AGV)used in the intelligent storage industry can replace the manual picking,access,handling and other work,which largely improves the operational efficiency of intelligent storage.However,in the environment where multiple AGVs work at the same time,conflicts and deadlocks usually occur during their operation.At present,operators of mainstream unmanned warehouses in China usually adopt the method based on mixed area restriction to avoid the problem,but this processing method will lead to a significant decrease in the overall efficiency of AGVs and also make AGVs go around,which has a great limitation on the operation efficiency of multiple AGVs in unmanned warehouses.The purpose of this paper is to propose an effective traffic control strategy to ensure that the system is not paralyzed by the occurrence of collision and deadlock under the premise of ensuring a higher operational efficiency of AGVs.In the paper,Dijkstra-based path finding and genetic algorithm-based task assignment improve the cooperative working efficiency of AGV to a certain extent,while the deadlock recovery algorithm based on time window,based on the way of changing paths and waiting,proposes a new deadlock recovery method of correcting and transferring path resources according to the occupation time of resource points to ensure that the deadlock dilemma can still be lifted in complex road networks.Unlike the traditional design usually used in the mainstream unmanned warehouse in China,the deadlock detection algorithm proposed in this paper starts from the current single AGV and only detects AGVs that conflict with the current AGV resource occupation time point,while generating a global time window cache in the upper-level scheduling system to avoid conflicts among AGVs at the global level while retaining the autonomy of individual AGVs,which is suitable for use in large intelligent warehouses.It is suitable for use in large intelligent warehouses.