Two-stage Multi-AGV Path Planning Based on Speed Pre-allocation

摘要

Dynamic response speed and energy consumption have always been important indicators for evaluating the advantages and disadvantages of automatic guided vehicle (AGV) dispatching systems. In order to reduce the energy consumption of multi-AGV path planning system, improve its working efficiency and promote the efficient operation of automated production workshop, this paper proposes a two-stage multi-AGV path planning method for speed pre-allocation. In the offline phase, the Dijkstra algorithm is used to generate K shortest paths for nodes with task associations, that is, the path library is generated offline. In the online phase, firstly, according to the task start and stop nodes of each AGV, the shortest path set with no relative conflict is found, and then the priority of the AGV is dynamically assigned according to the number of node collisions and the same number of nodes between the paths of each AGV. Finally, the priority speed and the node timetable are used to pre-allocate the speeds of the AGV segments to obtain the link speed window, and generate an AGV operation scheme with fast response speed and low energy consumption. The simulation results show that the improved algorithm can reduce the total energy consumption of the system while ensuring the fast response of the system, and can effectively avoid collision conflicts, which greatly increases the flexibility and robustness of the AGV scheduling system.