Research On Scheduling Strategy Of Multi-AGV System

Automated guided vehicle(AGV)is a means of transportation that can move autonomously in accordance with a prescribed path under unmanned conditions.In the relevant research of AGV,solving the deadlock conflict in the multi-AGV system is the key to realize the cooperative operation of multiple AGVs.This article focuses on this problem.This article first studies the problem of deadlock avoidance in open systems.Banker's algorithm is a classic deadlock control algorithm with polynomial time complexity,but it is a suboptimal algorithm.The open system state security judgment algorithm is an optimal algorithm,but it has exponential time complexity in general,but there are also studies that show that the open system state security judgment algorithm has polynomial time complexity in some special graphs.Therefore,for practical applications,performance analysis of these two algorithms is of great significance.This chapter first studies the realization of these two algorithms in the AGV system.Based on this realization,it analyzes the effect of Banker's algorithm compared with the optimal algorithm to provide a basis for practical application.In addition,in terms of efficiency,this paper also studies the performance of the open system state security judgment algorithm on different graphs,especially on those graphs with actual characteristics.The research results provide the design of the AGV roadmap guide.Then,considering that there is an area with unlimited capacity in the open system,it cannot be directly used in actual projects.Therefore,this paper studies the problem of deadlock avoidance in closed systems and proposes a deadlock control algorithm based on graph structure analysis.First,establish a mathematical model of the system,and use a partially directed graph to describe the state of the system.On this basis,the chain structure and capacity in the figure are analyzed.The chain is a specific subgraph structure,and the capacity reflects the number of AGVs that can be accommodated in the sub-graph without deadlock.Existing studies have shown that the chain is an important structure for judging whether the system state is safe or not,and the capacity of the subgraph in the graph is closely related to the chain structure.Existing research has proved that in the reachable state of the system,if there is a state whose graph description has a chain structure,then the current state is safe(that is,it will not deadlock).This paper proposes a strategy to ensure that the system will not deadlock by maintaining a reachable state with a chain structure.This algorithm can effectively avoid the deadlock problem in the actual multi-AGV system.Finally,based on the above research,the AGV scheduling control software is developed,and the effectiveness of the above algorithm is verified through testing.