In recent years, the burden per worker has increased due to a decrease in the working population. Wearable assist suits have been developed as one of the methods for solving the problem. To extend the assist suit to practical situations, it is necessary to provide a motion judgment interface for judging the motion of a wearer. Therefore, in our study, a motion judgment algorithm is proposed for assist suits, based on variable viscoelasticity. The proposed algorithm judges sitting, standing-up, stance, sitting-down, and gait using only the joint angle information of the suit and verification is performed using human joint angles obtained by motion capture. Thus, the motion judgment rate is 90% or more for sitting, standing-up, stance, and sitting-down, and 80% or more for gait, confirming the usefulness of motion judgment. Additionally, based on these results, further verification is performed on an actual machine. As a result, in a series of motions starting from the sitting to the standing-up, the stance, and the gait, the motion judgement is successful five times from the sitting to the standing-up, the stance, and once in gait. In a series of motions from sitting to standing-up, the stance, and sitting-down, the motion judgment is successful five times during sitting; five times during sitting, stance, and sitting-down; and three times during standing-up. In this way, it is confirmed that the proposed method can judge the motion only by angle information, although there is a problem in a success rate depending on the motion.
Determining the Online Measurable Input Variables in Human Joint Moment Intelligent Prediction Based on the Hill Muscle Model
