Sorting Multibay Block Stacking Storage Systems with Multiple Robots

Swarms of autonomous mobile robots are increasingly used to automate operations in intralogistics. One pivotal feature of robots is their availability, enabling them to optimize the allocation of unit loads in idle times. This work addresses the multi-robot multibay unit load pre-marshalling problem. The problem aims to reshuffle a warehouse in preparation for future demand while using multiple robots. The usage of multiple robots requires tour planning between reshuffling moves while respecting dependencies between the moves. This paper compares an interlaced approach and a sequential approach. The interlaced approach combines reshuffling and tour planning in a tree search. The sequential approach first uses a tree search to identify short reshuffling moves, then determines move dependencies, and finally plans robot tours between the reshuffling moves. We investigate two objectives for the minimal number of reshuffling moves: makespan and travel time. The results demonstrate that the sequential approach can achieve similar or better results in a shorter solution time.