Research On Path Planning Algorithm And Collision Avoidance Method For Grid Map Based Multiple Mobile Robots

With the increasing flexibility and intelligence of mobile robots,intelligent logistics handling systems based on multiple mobile robots are increasingly used in intelligent warehousing and port transportation.How to improve the balance of the path distribution of multi-mobile robots and avoid collision and deadlock phenomenon of the multi-mobile robot system is the most important to ensure the operational efficiency and stability of multi-mobile robot system.This paper investigates the critical technologies of path planning and collision avoidance traffic management for a multimobile robot system applied to courier sorting.The main work includes the following parts.Firstly,a system architecture is designed for a multi-mobile robot system used in courier picking;secondly,the raster method is used to model the multi-mobile robot picking operation area after comparing several commonly used environment modelling methods;finally,the operation flow and navigation and positioning methods of individual mobile robots are designed for the multi-mobile robot picking operation requirements.This method first identifies the temporarily allowed running directions of each path according to the planned path library,ensuring that all paths are unidirectional and avoiding deadlock caused by conflicting paths between two robots.Secondly,the path search process takes into account not only the distance travelled but also the traffic flow and the number of turns at each grid point,thus improving the balance of the path distribution of the multi-mobile robots and avoiding local congestion in the path network,thus improving the overall efficiency and intelligence of the multi-mobile robot system.A collision avoidance and deadlock prevention method based on dynamic allocation of grid points is proposed for multi-mobile robots used in courier sorting.First,several common motion conflicts in multi-mobile robot systems are addressed,and corresponding motion conflict resolution strategies are proposed.The collision between multiple mobile robots is avoided.Secondly,for the loop deadlock prevention problem,a forward-looking one-step strategy is adopted,and a loop deadlock search and avoidance method based on directed graph theory is proposed.Finally,a multi-mobile robot collaborative system for express parcel sorting is used as an example,and a simulation model of the multi-mobile robot collaborative sorting system is constructed.The analysis results show that,compared with the classical Dijkstra algorithm,the multi-objective optimised path planning algorithm proposed in this paper can significantly improve the balance of the multi-mobile robot's path distribution picking efficiency of the multi-mobile robot system.The analysis results also show that the proposed collision avoidance traffic management method can avoid collision between multiple mobile robots and effectively avoid the loop deadlock phenomenon,which ensures the stability and reliability of the operation of the multimobile robot system.