大規模言語モデルを用いた自律協調移動ロボット群のタスク割当制御の提案と概念実証

抄録

Autonomous mobile robots employ various cooperative control methods, including consensus control, coverage control, and formation control, to achieve global objectives. However, a key issue is that only the robots themselves adapt to control tasks based on environmental conditions and cyber-physical constraints. To address this limitation, this paper proposes a novel human-robot control architecture that leverages human-guided individual task assignment via large language models to assign control roles to specific robots. A proof of concept for the proposed control architecture is demonstrated through simulations and voice-prompt-based experiments.