Dynamic Scheduling Strategy for Multiple AGVs based on 0-1 Planning Model

Abstract

The coordination of handling robots and AGVs can greatly improve the efficiency of unmanned warehouse picking operations and realize the automated management of warehouse logistics. The unreasonable scheduling scheme can lead to local congestion and collision of AGVs and reduce the picking efficiency of the unmanned warehouse. Therefore, this paper conducts an in-depth study on the scheduling mode of AGVs. When assigning orders, this paper designs an order-oriented double matching method, with the shortest total path length of AGV as the objective function, and establishes a 0-1 planning model to complete the matching of order-storage position nodes-AGV. The AGV searches and selects the nearest picking station node whose docking position is not full when it leaves the warehouse. In addition to the main order list, a workstation task list is created, and the workstation tasks are released by the picking stations, including return and recovery. In this paper, the default order and workstation tasks have the same priority, and the task selection order is the same as the release order. The 0-1 planning model is solved using a genetic algorithm to match the AGV with the order or station task that has the shortest path to completion.