Optimal Path Planning for a Robot Shelf Picking System

摘要

In this paper, we concern ourselves with the development of a robot system to improve the efficiency of shelf picking. A 6-degree-of-freedom robot is deployed to detect, grasp and place objects on the shelf according to order information. To generate the order picking sequence with the shortest path, an optimal path planning approach is proposed. The picking path planning is abstracted as a Traveling Salesman Problem (TSP) with sequence constraints and path form constraints. An improved hybrid genetic algorithm is presented to optimize the picking path. The experiment shows that the proposed method can shorten the picking path and improve efficiency compared with traditional picking methods.