Research On Task Assignment And Path Planning Of Multi-Mobile Robots Based On Intelligent Warehousing

With the popularity of the Internet and continuous technological advancements,the ecommerce industry has shown a booming trend in the past few years.The rapidly growing logistics volume has also put higher demands on the logistics industry.In recent years,more and more e-commerce platforms have established their own intelligent warehouse systems in major cities,replacing manual handling of goods with mobile robots.This not only increases the throughput of the warehouse but also reduces the risk of employee injury.To further improve the efficiency of intelligent warehouses,this paper mainly focuses on the task allocation and path planning of multiple mobile robots.While ensuring that mobile robots can complete tasks more quickly,it also ensures that no conflicts occur during task execution.The main research content is as follows:(1)A single-objective task allocation model is established based on the shortest total distance traveled to complete all tasks for the problem of multiple mobile robot task allocation.An improved genetic algorithm-based multiple mobile robot task allocation algorithm is proposed.Firstly,by changing the encoding method of the genetic algorithm,the sequence of tasks that each mobile robot needs to perform is better represented.Secondly,by changing the genetic operation of the genetic algorithm,the problem of producing solutions that do not meet the constraint conditions is solved.Simulation experiments show that using the improved genetic algorithm proposed in this paper for task allocation,the algorithm converges faster,and the total cost consumed by the system is smaller.(2)For the problem of single mobile robot path planning,this paper selects the A* search algorithm based on the Manhattan distance as the basic algorithm for path planning.In response to the problem of too many turning points on the path obtained by the A* algorithm,a turning penalty factor is introduced to reduce the number of turning points on the path.In response to the problem of too many search nodes and long algorithm execution time during the A* algorithm search process,the heuristic function of the A* algorithm is dynamically weighted to reduce the number of search nodes during the search process and speed up the execution efficiency of the algorithm.Experimental results show that the improved A* algorithm reduces the number of turning points by 75.26%,the number of search nodes by 50.23%,and the algorithm execution time by 47.37%.(3)For the problem of multiple mobile robot path planning,a centralized planning-based conflict search algorithm is used,which divides the path planning of multiple mobile robots into two levels for solving.In the high-level search,the algorithm mainly solves the conflict problem on the path by adding constraint conditions for mobile robots that have conflicts to solve the conflict.In the low-level search,the path planning for multiple mobile robots is carried out independently,without considering other mobile robots,and ultimately obtains a path that does not conflict with other mobile robots.