Research On Dynamic Optimization Of AGV Scheduling System

AGV(Automated Guided Vehicle)is a wheeled mobile robot,which has the advantages of fast driving speed,simple structure and high safety performance.In the future automated logistics and even unmanned factories,AGV is very important,and its reasonable use will make factory production more flexible,flexible and efficient.The multi-AGV scheduling system is connected to the upper system(ERP,WMS,etc.),and the task order is issued to the AGV through the internal algorithm.The dispatching system monitors the running status of the AGV in real time,so that the AGV can cooperate with other equipment to complete the task together.With the advancement of Industry 4.0,how to perform efficient AGV scheduling has become a research hotspot.This paper analyzes and summarizes the current research status of AGV dispatching systems at home and abroad,and has a deeper understanding of it.On this basis,the secondary development based on openTCS open source dispatching system software can achieve a more efficient and stable dispatching plan.To avoid the occurrence of deadlock phenomenon.The research content of this article is as follows:The software and hardware of the whole dispatching system are analyzed and selected.The dispatching system communicates with the AGV,and the dispatching end system directs the movement trajectory of the AGV.In addition,there are edge devices(such as workstations,charging piles,etc.)that communicate with the dispatching system.The dispatching system serves as an intermediate control processing unit,making the AGV and other There is an indirect connection between the devices.In addition,this article models the factory model.Due to the complexity of the processing on the dispatch system,in order to ensure its real-time performance,the topology map based on graph theory is used.To summarize and analyze the advantages and disadvantages of existing path planning algorithms,task allocation algorithms,and traffic control algorithms.On this basis,an improved adaptive genetic algorithm is proposed for the path algorithm.The algorithm introduces inversion operators,insertion operators and catastrophe operators.For the task allocation algorithm,order tasks are allocated according to priority.For the traffic control algorithm,a deadlock judgment mechanism is established.When a conflict occurs,it is sorted according to priority,and then queued.If a deadlock occurs,re-plan the path according to the deadlock judgment mechanism.According to the modified path algorithm,task allocation algorithm and traffic control algorithm,scheduling can be effectively carried out,conflicts can be avoided as much as possible,and the deadlock problem can be effectively solved.Based on openTCS open source software,complete secondary development.It communicates with other devices and systems through wireless local area networks.Communicate with charging piles and AGVs to realize the automatic charging function of AGVs;develop orders from file reading to release orders,which can carry out transportation tasks in large quantities;develop ERP docking interfaces,accept ERP information,and perform corresponding scheduling tasks;develop web pages Version interface,you can monitor the operation of AGV in real time on the web page,view historical data,release scheduling tasks,etc.Based on MATLAB and openTCS simulation analysis and path algorithm,it can be found that the improved genetic algorithm has faster convergence and better solution results.In terms of traffic control,path re-planning can be carried out effectively to avoid deadlocks.On the whole,the improved algorithm achieves better results and better results.