Abstract
The man-machine cooperative system is attracting great attention in many fields. Since a human skill is strongly affected by fatigue, the assist system must be designed so as to accommodate with the change of skill characteristics caused by fatigue. This paper presents a new fatigue recognizer based on the EMG signals and the stochastic switched ARX (SS-ARX) model. Since the SS-ARX model can represent complex dynamics which involves switching and stochastic variance, it is expected to show higher performance as the fatigue recognizer than using simple statistical characteristics of the EMG signal. The usefulness of the proposed strategy is demonstrated by applying to a peg-in-hole task.
