Research On Scheduling Optimization And Collision Avoidance Strategy Of Warehouse Multi-AGV System

In recent years,the rapid development of online e-commerce has accelerated the innovation of the logistics industry.Smart logistics is gradually replacing traditional logistics and automated guided vehicles(AGVs)occupie an important position in it.As a key node of smart logistics,the warehousing multi-AGV system directly determines the competitiveness of the enterprise,and it is of great significance to improve the system scheduling optimization and collision avoidance capabilities.This paper takes the "part-to-picker" picking mode in the warehouse multi-AGV system as the research object,and focuses on the multi-AGV scheduling optimization method and collision avoidance strategy.The specific research work is as follows:(1)In the path planning of warehouse AGV,in view of the problems of many turns in the path planned by the traditional A* algorithm,an improved A* algorithm that reduces the number of turns is proposed,and the concept of "AGV direction sense" is introduced.By judging and comparing the target position and the AGV position,and combining the obstacle matrix,replace the heuristic function in the traditional A*algorithm with the "direction" function.The AGV can predict the obstacle and find another path when expanding the path node.(2)Based on the AGV path planning method,the scheduling optimization problem of the warehouse multi-AGV system is analyzed.To solve the problem of task allocation and sorting in static scheduling,an optimization model is established based on the shortest total AGVs travel and the shortest total task completion time,and a GAACO(Genetic Algorithm,Ant Colony Algorithm)two-stage optimization algorithm is proposed.To solve the problem of task allocation and sorting in dynamic scheduling,the AGV traffic volume of each road section in the warehouse in the short term is balanced.An optimization model is established to reduce the number of conflicting nodes in the path and the total travel time of the AGVs is the shortest,and a dynamic task allocation and sorting strategy based on the short-term state simulation of the system is proposed to solve the problem.(3)By analyzing the conflict types that occur during the operation of the warehouse AGVs,a conflict detection method based on real-time information is proposed,and corresponding collision avoidance strategies are adopted for AGVs that may conflict according to the conflict types.For intersection congestion,a dual collision avoidance strategy is formulated: by setting up detection and buffer zones at intersections,a traffic sequence optimization method is proposed;if this method cannot solve intersection congestion,a local path planning strategy is adopted to ensure smooth passage of all AGVs.(4)A warehouse multi-AGV system experimental platform was built,and the improved A* algorithm,static scheduling method,dynamic scheduling and collision avoidance strategy proposed above were tested experimentally,and their feasibility and effectiveness were verified respectively.