Research On Multi-AGV Path Planning And Task Scheduling

With the advancement of robotization and intelligence,the status and value of automatic guided vehicles(AGVs)are becoming more and more important,because AGVs are widely used in logistics,warehousing and other industries due to the characteristics of high degree of automation,low production costs and high working efficiency,so the investigation on AGV’s key technology path planning and task scheduling is extremely important.Firstly,establish electronic map model before AGV planning based on storage environment,analyze common electronic map modeling methods and then select raster map to complete subsequent simulations according to the advantages and disadvantages of electronic map.Secondly,for the problem of AGV path planning,first of all,study on single AGV path planning,Analyzes the advantages and disadvantages of common AGV algorithms,and then select A* algorithm,Aiming at the shortcomings of traditional algorithms,an improved A* algorithm based on dynamic weights and two-way search mechanism is proposed,By comparison,analyzes the advantages of the improved A* algorithm in contrast to the traditional A* algorithm,and then aiming at the conflict and collision between AGVs in multi-AGV path planning,propose an improved A* algorithms based on time windows.The time windows records the position information of the AGV at each moment,and determine if there is conflicts,finally,according to the conflict resolution strategy to resolve the conflict,and then tests the practicability of the algorithm in simulation on Matlab.Thirdly,for the task assignment problem of multi-AGV,an modified genetic algorithm is raised,based on the original genetic algorithm,proposes local search optimization to avoid falling into local optimal solution too early.According to the goal of the shortest overall running distance of the AGV,compared with traditional genetic algorithm,we verifies the shortening of distance with improved algorithm on Matlab,and verifies the improvement of overall system by improved genetic algorithm.Finally,based on the ROS operating system,the relevant experimental verification of AGV is carried out.