Research On AGV Path Planning In Intelligent Warehousing Environment

Intelligent warehousing is an important part of intelligent logistics system,and its intelligent level also reflects the development status of modern logistic.At present,AGV scheduling and route planning has become a research hotspot in intelligent warehousing.Through the reasonable scheduling of AGV in the warehouse environment,the turnover efficiency of materials can be effectively accelerated and the logistics cost can be reduced.In this thesis,multi-AGV task scheduling and path planning in intelligent storage environment are studied.The main research contents are as follows:(1)Research on Single AGV path planning in intelligent warehouse environment.Firstly,based on the intelligent warehouse environment studied in this thesis,grid method is used as the environment modeling method.Secondly,RRT*algorithm is selected for path planning of single AGV.Aiming at the shortage of the traditional RRT* algorithm with a large number of traversal node,this thesis puts forward to join the artificial potential field method improved RRT * algorithm to accelerate the search speed of the path.And under the environment of smart bookstore the AGV path planning simulation experiment,the results show that the improved RRT* algorithm has fewer traversal nodes and shorter path optimization time.(2)Research on multi-AGV task scheduling in intelligent warehouse environment.In this thesis,the centralized control system is selected as the total control system of multi-AGV task scheduling.The objective function of multi-AGV task scheduling is the shortest AGV running time.This thesis proposes an improved genetic algorithm with natural number coding,multi-point crossover and single-point mutation.It is applied to the task scheduling in a warehouse workshop environment and verified by simulation.The results show that the improved genetic algorithm has fewer iterations and shorter time than the traditional genetic algorithm.(3)Research on multi-AGV path planning in intelligent storage environment.Based on the conflict types and collision avoidance strategies of multi-AGV,the A*algorithm and time window algorithm are analyzed.In order to solve the problem that the bidirectional search A* algorithm searches too many neighborhoods,this thesis changes the traditional 8-neighborhood search to 5-neighborhood search,and adds Bezier smooth curve to reduce the number of inflection points of the path.The simulation of multi-AGV path planning was carried out in the printing workshop environment.The results show that the improved bidirectional search A* algorithm based on time window can better realize the path planning of multi-AGV without collision and few inflection points.