Conflict Algorithm Design And Simulation System Implementation Of Multi-AGV Path Planning

The paths mapped by Automated Guided Vehicle(AGV)missions are of great significance for industrial scenarios such as warehouses,logistics sorting and unmanned Automated factories.In recent years,with the rapid development of the logistics and transportation industry and the increasingly urgent requirements of industrial automation,the requirements of automatic storage system for material transportation are increasingly high,and AGV is an important part of such automatic storage system.In order to improve transportation efficiency,people began to study how to transport as many goods as possible in a given time,in which the collision of AGVs with each other is the key factor of time consumption.The goal of this article is to ensure that all AGVs in the scenario have the least amount of conflict with each other while completing the task.In order to solve the problem of multi-AGV high dynamic and less conflict,the following aspects of work were mainly completed:Aiming at the static scenario algorithm research of multi-AGV,this paper proposes an improved heuristic pathfinding algorithm to enhance conflict avoidance.After investigating the layout characteristics of logistics storage site,based on the raster network model,the storage scene map was abstracted into the form of undirected graph,and the tasks performed by AGV were described in detail according to the characteristics of static scenes.Based on the idea of cooperative path planning,an improved heuristic search scheme was proposed,which significantly improved the number of path turns,overlap degree and the number of AGV conflicts,and greatly reduced the indexes required by AGV to complete tasks.Aiming at the research of multi-AGV dynamic scene algorithm,this paper innovatively designs and implements a high-dynamic,conflict-oriented algorithm suitable for specific problems.The characteristics of dynamic algorithm are analyzed and the research idea of dynamic algorithm is determined.For the first time,a concept of pheromone matrix is proposed,which is suitable for dynamic scenes.The algorithm can greatly reduce the algorithm execution time and avoid the secondary conflict caused by vehicle following.By studying the theory of partial node expansion,a generalization algorithm based on the theory is proposed,which makes the speed of the basic path planning stage significantly higher than that of the comparison algorithm.In static algorithm used in thefeature of the global reservation table,innovation in this paper proposes a dynamic request table structure is suitable for high dynamic scene,in view of the actual problem of conflict resolution strategies and the concept of dynamic priority,not only well realize the dynamic nature of the algorithm,but also let the task execution process of conflict evenly distributed between the AGV,through contrast experiment proves that the effectiveness of the algorithm on conflict prevention.After completing the theoretical construction of the dynamic algorithm,this paper designs and implements the simulation system according to the actual needs of schoolenterprise partners.In the absence of a feasible solution strategy,the problems encountered in the transportation of production materials were solved one by one.At the same time,the functions are designed and integrated to build a set of simulation system,which has been verified and affirmed in the enterprise,and realized the intuitiveness and interactivity of the research content,and embodied the practical engineering value of the actual production.