Research On Cooperative Scheduling System Of Storage Logistics Robot Cluster

Warehouse is the center of logistics operation.With the rapid development of logistics industry and the rapid rise of e-commerce,the scale of warehouse is expanding continuously,it has the characteristics of miniaturization,many varieties,small batch,more batches,short cycle and so on.It also leads to the high-frequency update of warehouse data operation.The information structure is not only complex but also varied and the task of commodity selection is heavy.Traditional warehousing systems,such as manual operation,conveyor belt or AGV,are no longer sufficient to deal with the current problems,such as informatization,modernization and systematization of management systems,and rationalization of storage management.More and more attention has been paid to the trend of humanized development.People urgently need a new type of intelligent automatic warehousing logistics,which has low error rate and low labor cost,and improves the whole operation efficiency of the storage system at the same time.In this context,the intelligent warehousing system,which uses the warehousing logistics robot to replace the manual picking work,emerges as the times require.In the intelligent warehousing system,how to effectively schedule the warehousing logistics robot so that it can efficiently complete the order task,It has become an important issue whether intelligent storage system can finally be realized.Therefore,based on the warehouse logistics robot cluster,the collaborative scheduling problem in intelligent warehousing is studied and analyzed in this paper.Firstly,a flexible and reconfigurable rasterized warehouse model is constructed.On the basis of this model,the scheduling problem is divided into two parts: task assignment and path planning.In this paper,a multi-agent task assignment algorithm based on multi-agent coding genetic algorithm considering time cost,path cost and coordination cost is proposed,and the path planning scheme of agent is realized by using the Q-Learning algorithm in reinforcement learning.By setting up appropriate reward and punishment mechanism,the convergence of Q-Learning algorithm is accelerated and the accuracy is improved,and the structure of the algorithm is optimized by using Manhattan path to estimate the cost of path.The performance of MATLAB simulation results is improved by more than 20% compared with the related literature results,which is more suitable for solving complex large-scale intelligent warehousing scheduling problems.The obstacle avoidance of robot cluster in path planning and the cooperative operation in work are the key to the normal operation of intelligent storage system and the improvement of its operation efficiency.In this paper,a new method of obstacle avoidance and cooperative operation under the constraints of traffic rules and reservation tables is proposed.Through the path planning algorithm,the shortest path of each robot to accomplish the task goal is planned and the reservation table is formed.The collision and deadlock of the storage logistics robot cluster are solved by using the traffic rules and the reservation table,and the problem of collision and deadlock occurs in the storage and logistics robot cluster is solved.According to the coordination mechanism,the task-free standby state of the robot is reduced,the workload of each robot is balanced,and the aim of shortening the running time of the system is realized on the basis of ensuring the safe operation of the system.The algorithm designed in this paper is simulated by MATLAB,and the total running time of the system to complete all tasks without collision,that is,the total number of steps to complete the last task in the system,is taken as the evaluation index,and the effectiveness of this method is verified.