Research On Multi-AGV Scheduling And Path Planning Of E-commerce Logistics Sorting Center

As an important hub of express transportation,the sorting efficiency of logistics sorting center affects the delivery time of express to a certain extent.With the rapid development of major e-commerce,the number of packages increases year by year,and the logistics sorting center puts forward higher requirements for the efficiency and flexibility of sorting solutions.Traditional manual sorting has been unable to meet the needs of modern sorting,cross-belt sorting system has certain limitations on the scale and applicable commodities.Automatic Guided Vehicle(AGV)as an automated modern logistics equipment with flexibility,efficiency and cost advantages,has been more and more widely used in the logistics sorting center.In the AGV sorting system,the scheduling scheme affects the sorting efficiency of the system,so the task scheduling and path planning of multiple AGVs has become an urgent problem to be optimized in the current logistics enterprises.This paper studies the multi-AGV scheduling problem of E-commerce Logistics Sorting Center.By analyzing the AGV sorting operation process and demand characteristics,it explores the efficient problem solving ideas.Because of its NP-hard(Nondeterministic Polynomial-Hard)property,it is often difficult to find the optimal solution in a reasonable time.Therefore,a corresponding optimization strategy is designed based on constrained programming and neighborhood search.The main research contents of this paper are as follows:(1)The development history of logistics sorting and AGV at home and abroad is analyzed,and the research status of multi-AGV scheduling problem and path planning problem at home and abroad is summarized and sorted,so as to provide technical reference for the research of the following problems.(2)Considering the power consumption and charging demand of pure electric driven AGV during transportation,the multi-AGV task scheduling problem considering charging demand and time window was studied.And a mixed integer programming model was established to minimize the completion time of sorting.By analyzing the characteristics of handling tasks,two constraint programming models were established.The results of an example were used to verify the effectiveness of the model.Based on constraint programming technology,the constraints of the problem are extended and analyzed to be applicable to more practical scenarios.(3)The general variable neighborhood search algorithm is used to solve the large-scale scheduling problem.Based on the algorithm framework,a heuristic method is designed to obtain the initial solution by combining the characteristics of the problem,and ten different neighborhood structures are added.The results show that the proposed algorithm has advantages in computational time and performance,and the effects of different AGV charging rates and number configurations on the sorting efficiency are analyzed.(4)The problem of multi-AGV conflict-free path planning for logistics sorting center is studied.Combined with the scene features,the raster map method is selected to conduct environmental modeling.A* algorithm,which performs best in sorting scenarios,is proposed to be introduced into the time dimension as a low-level planer based on conflict search.The ability of the improved algorithm to resolve different conflict situations is verified by simulation experiments.Aiming at the multi-AGV sorting process of E-commerce Logistics Sorting Center,this paper considers factors such as AGV charging demand,package sorting time window and AGV conflict,and solves the corresponding problems by establishing mathematical model and design algorithm.The research in this paper is helpful to improve the efficiency of logistics sorting,enrich the research content of multi-AGV scheduling and path planning,and has certain theoretical and practical significance.