「手応え関数」に基づくミミズ型ロボットの自律分散制御

抄録

Although autonomous decentralized control could be the key to understand the mechanism of adaptive and resilient animal locomotion, its mechanism has not yet been clearly understood. To tackle this problem, we here focus on earthworms, which exhibits crawling locomotion by using their simple one-dimensional body structure. We designed a decentralized control scheme for earthworm locomotion on the basis of TEGOTAE function, which determines cause-and-effect relationship between action and reaction. Owing to sensory feedback based on TEGOTAE function, each segments can selectively exploit reaction force that assists propulsion in real time. We performed simulations and then verified the adaptability, fault-tolerance and scalability for the proposed control scheme.