Research On Multi-AGV Job Scheduling Based On Improved Ant Colony Algorithm

Multi-AGV operation is becoming more and more widely used as the scale of industry expands,the degree of automation increases,the number of processes increases and the flexibility increases.The job scheduling problem of the multi-AGV can produce certain effect on improving transportation efficiency,therefore,it is of great significance to research it in the practical application of rich multi-AGV operation scheduling.This thesis focuses on the path optimization problem of AGV in the process of loading parts of the workshop,multi-AGV scheduling problem in intersection collision avoidance,established the mathematical model of AGV to complete the task driving shortest path as the objective function,and proposed an improved ant colony algorithm to solve the model.The main research contents of this thesis are as follows:(1)On the basis of studying the relevant literature at home and abroad,according to the basic ant colony algorithm iterative convergence is slow,easy to jump out of the global only in the local expansion of the optimal solution search and other defects,an improved ant colony algorithm is proposed to update local and global pheromones simultaneously,and gives“the maximum and minimum cross strategy”,obtain a new pheromone update scheme,to improve the traditional algorithm.(2)Combined with multi-AGV workflow,with the help of the traffic flow distribution of the traffic impedance thought,consider the work path on the roads and intersections and secondary paths,respectively from the AGV job scheduling of traffic demands,the parameters of speed,driving path three aspects of research.The working area of AGV walking in the experimental field is analyzed and transformed into two-dimensional structured space.The environment map of AGV walking space is established by using the grid identification method of direct coordinate to prepare the later simulation experiment.(3)For multi-AGV collision avoidance at the intersection of the problem,put forward the high and low speed reduction adjust the strategy of solving intersection conflict,and established the mathematical model of AGV to complete the task driving shortest path as the objective function,and the improved ant colony algorithm is used to solve the problem,using Matlab software for ant colony algorithm and the improved ant colony algorithm in simulation experiment,and analysis of the experimental data,the results show that the improved ant colony algorithm under spend less time cost can continue to overflow local optimum,to find the new optimal path in the global region,and the optimal solution,the average and worst solutions have higher precision and better performance.The validity of the improved ant colony algorithm in multiAGV scheduling optimization is verified,which can provide reference for the actual scheduling operation of AGVS.(4)In order to improve the degree of automation of workshop parts transport,to achieve the minimum input to maximize the benefits of the target.In this thesis,VB6.0 for the development of the system platform,SQLserver2008 for the background database,with Matlab software,design and development of multi-AGV scheduling management system,this system not only realize the function of parts of loading and unloading automation management,at the same time,the improved ant colony optimization algorithm written in Matlab is called in VB6.0 by using mixed programming,and the order of the parts loading is given.The optimal scheduling and management of the multi-AGV dispatching management system is realized,and the efficiency of AGV in the workshop is improved.