Research On The Key Technologies Of Dynamic Path Planning In Multi-robot For Intelligent Warehouse

As the rapid development of e-commerce and industrial automation,the demands for intelligent automated warehouses is growing.At the same time,artificial intelligence has become a national development strategy,and intelligent robots have become the developing direction of the field of artificial intelligence research.Therefore,the study of intelligent warehouse systems based on multi-robot has become a hotspot in the intelligent manufacturing industry.This paper presents a complete intelligent warehouse system based on multi-mobile robots,and proposes solutions to the problems such as cooperative path planning,wall fault and task collating between robots.First of all,this paper proposed a hybrid control system to control multi-robots system.That is,central control system deals with the information such as position,status and task allocation in a centralized way,and robots complete path planning in a decentralized way.Secondly,an improved A*algorithm based on traffic rules,reservation tables and dynamic weighted map was proposed for multi-robots cooperative path planning.Traffic rules and reservation tables can solve the collision and deadlock problem between multi-robots,and the dynamic weighted map can solve the traffic congestion problem in warehouse.Finally,we proposed the genetic algorithm to sort the order tasks that the robots will perform,which can reduce the driving path and improve the working efficiency of the robots in advance.At the end of this paper,we used MATLAB to simulate the intelligent warehouse system that proposed in this paper.And through the comparison experiment,we verified that the collaborative path planning method and order sorting method can effectively improve the efficiency of intelligent warehouse system.