Mutual-entrainment-based internal control in adaptive process of human-robot cooperative walk

Abstract

Mutual-entrainment-based internal control is proposed to overcome the limitation of the control theory which is based on the separation between the system and its environment. In this new approach, since the system's dynamics is inseparable from that of the environment, it should be composed of two different activities. One self-organizes the relationship between the system and the environment by using "mutual entrainment" as a non-linear interaction, and the other controls the organized relationship based on that relationship. This relationship-based internal control is realized as a mutual adaptation process in an autonomous robot which walks cooperatively with humans.