Research On AGV Optimal Configuration And Two-stage Path Planning Algorithm For Unmanned Warehouse

In recent years,with the vigorous development of the Internet industry,online shopping has been rapidly into the thousands,traditional manual sorting goods have been difficult to meet the needs of users and the development of the electricity,the emergence of the "goods to people" sorting model and the establishment of the unmanned storehouse brought dawn to electricity,the research of some important factors affecting unmanned warehouse operation efficiency,how to improve the sorting efficiency of unmanned warehouse has become the current academic and engineering problems need to be solved.Unmanned warehouse is a comprehensive system,which contains multiple operation parts.Among them,the number of AGVs put into operation and the operation path of AGVS are two important factors affecting the efficiency of unmanned warehouse.There are many factors affecting AGV quantity decision in unmanned warehouse,so it is difficult to predict the result by establishing simple linear model.In terms of AGV path planning,some scholars proposed a two-stage path planning algorithm in combination with the characteristics of unmanned warehouse AGV operation path.Although this algorithm avoids the waste of multiple path planning for the same point(starting point and end point),it does not consider the efficiency reduction caused by AGV blocking or secondary blocking for unmanned warehouse.How to decide the number of AGVs put into unmanned warehouse and how to reduce AGV blocking degree as far as possible,so as to maximize the sorting efficiency of unmanned warehouse.Specifically,the main work of this paper is as follows:Firstly,the process of unmanned warehouse is analyzed in combination with the research content of this paper,and the main framework of the subject is proposed.The scheduling system and path selection in the framework are analyzed and studied.Secondly,in view of the unmanned warehouse AGV decision making problems,the number of too much due to the factors affecting the number of AGV in unmanned storehouse and variability is high,so this article constructed the simulation optimization scheme for AGV quantity to make decisions,so that no one warehouse sorting efficiency maximization,in simulation optimization model,with high precision,fast convergence,good global particle swarm optimization(pso)algorithm as the optimization algorithm.Then,in view of the AGV path planning problem,according to most point in the unmanned warehouse don't need to repeat the characteristics of path planning,project path planning algorithm using two stage,collect affect the blocking of AGV path node,on this basis,respectively to the offline and online phases of path planning algorithm is studied and improved,the offline stage improved ant colony algorithm was proposed based on block node,online stage was proposed based on time axis direction the improved A * algorithm,at the same time for different priority of AGV assign A different path.Finally,through experiments to validate the above algorithm,the simulation optimization model for successful estimate the number of AGV,makes no warehouse sorting efficiency maximization,the improved ant colony algorithm program can be very good avoid blocking the path of node,the improved A * algorithm can reduce the AGV in secondary or multiple blocking probability,the whole two-phase path planning algorithm to A certain extent improve the warehouse sorting efficiency.