Research On Multi-robot Path Planning For Logistics Scenarios

The development of e-commerce has driven the sustainable and rapid growth of the logistics industry.With the inclusion of smart logistics into the national strategic development plan,the intelligent transformation of the logistics industry has become an inevitable trend.Using multi-robot system to replace traditional manual operation in warehouse environment can save logistics time and improve sorting accuracy,which has feasibility and superiority.Multi-robot path planning is the key to determine the efficiency and stability of the system.Each robot in the system needs to have the ability of autonomous navigation and collision avoidance,and multiple robots need to avoid conflicts in the process of moving through cooperative strategies.A reasonable multirobot path planning method can help logistics robots to obtain a path scheme that can balance system resource occupancy and task execution efficiency,which has research value and application requirements.In this thesis,the path planning method of multirobots in the warehousing and logistics scenario is studied as follows:(1)For the warehouse logistics scenario,a multi-robot path planning map model is established,in which unexpected obstacles exist in the dynamic environment map,so as to simulate the uncertainty of obstacles in the workspace of the robot in the intelligent warehouse environment.A multi-logistics robot path planning system is developed,in which the path planning module integrates the multi-robot path planning simulator for algorithm verification,and the task management and anomaly monitoring module simplifies the management and scheduling of intelligent warehousing.(2)Aiming at the global path planning problem of multi-robots,a multi-robot global path planning algorithm based on improved CA* is proposed.The improved algorithm can judge the waiting,turning and diagonal straight behavior in the global path finding process of multi-robot,and reflect the cost of cooperative behavior in the cost function.The simulation results show that the improved algorithm can improve the global path ordering and shorten the average path execution time.(3)Aiming at the problem of multi-robot local path planning,a multi-robot local path planning algorithm based on improved artificial potential field is proposed.The improved algorithm solved the local stagnation problem by obtaining the virtual target point through the migration cycle strategy based on the double trapping supervision.Accidental obstacle potential field based on expansion coefficient and multi-machine conflict potential field based on rotation Angle are used to improve the local collision avoidance ability of the robot.(4)Combining the advantages of the improved CA* algorithm and the improved artificial potential field method to avoid collision,the global and local path combination strategy is given,and a hybrid path planning algorithm for multi-logistics robot is proposed.The simulation results show that it can improve the execution efficiency of logistics path and reduce the collision risk of robot and unexpected obstacles in the warehouse environment.