Research On Task Allocation And Path Planning Algorithm For Multi-warehouse Robot Syste

In order to improve the overall operating efficiency of the multi storage-robot system to meet the storage needs driven by the development of e-commerce,two key issues affecting the efficiency of the storage system: multi-robot task allocation and path planning are absorbed in research.According to the actual operation requirements of the storage system,the optimization goal is defined,and the task allocation algorithm and path planning algorithm of the multi storage-robot system are mainly studied.First,the kinematics model of the storage-robot and the odometer-based method for estimating the trajectory of the storage-robot are established.By comparing three environment modeling methods,the grid method is finally used to model the storage environment.Next,the task allocation problem in the multi storage-robot system is studied.The storage retrieval tasks are the most critical tasks in storage system,their mathematical models for the tasks assignment problem are established by analyzing the key characteristics of them.The retrieval tasks allocation algorithm based on the market auction mechanism and the storage tasks based on staged optimization are proposed.Then,the multi-robot path planning problem in the storage system is studied.A CBS algorithm based on improved D*Lite is proposed to solve the global path planning problem of multi storage robots.The algorithm uses the D*Lite algorithm based on dynamic directional windows as the lower algorithm to provide the basic path for the storage robot.The upperlevel algorithm of it introduces a constraint tree to the basic path to resolve conflicts between multiple robots,so as to obtain the overall shortest and safe path planning scheme.Finally,an experimental platform for multi storage robots is built,and the corresponding management software is developed and designed.With the help of robot environment mapping and autonomous navigation functions,the designed task allocation and path planning algorithm of the multi storage-robot system are tested and verified,good results are achieved.