Dynamic Order Picking Method for Multi-UAV System in Intelligent Warehouse

For the logistics environment, multi-UAV algorithms have been studied for the purpose of order picking in warehouses. However, modern order picking adopts static order picking methods that struggle to cope with increasing volumes of goods because the algorithms receive orders for a certain period of time and pick only those orders. In this paper, by using the modified interventionist method and dynamic path planning, we aim to assign orders received in real-time to multi-UAVs in the warehouse, and to determine the order picking sequence and path of each UAV. The halting and correcting strategy is proposed to assign orders to UAVs in consideration of the similarity between the UAV's picking list and the orders. A UAV starts picking orders by using the ant colony optimization algorithm for the orders initially assigned. For additional orders, the UAV modifies the picking sequence and UAV's path in real time by using the k-opt-based algorithm. We evaluated the proposed method by changing the parameters in a simulation of a general warehouse layout. The results show that the proposed method not only reduces completion time compared to the previous algorithm but also reduces UAV's travel distance and the collapsed time.

Automated summary

This study aims to assign real-time received orders to multiple UAVs in the warehouse using the modified interventionist method and dynamic path planning, and determine the picking sequence and path for each UAV. A halting and correcting strategy is proposed to consider the similarity between the UAV's picking list and the orders. The results show that the proposed method reduces completion time, UAV's travel distance, and collapsed time compared to the previous algorithm.