Classification Method of Assistance Motions for Standing-up with Different Foot Anteroposterior Positions using Wearable Sensors

In this paper, we propose an educational system of assistance motions using wearable sensors. In addition, we will briefly introduce our proposed system and evaluate the classification method of assistance motion with different foot positions using wearable sensors as a first step toward developing an educational system. Our proposed system comprises shoe-type pressure sensors and a single inertial measurement unit attached to the trunk. The proposed method classifies assistance motion by machine learning technique using feature quantities obtained from wearable sensors. Herein we evaluated whether the proposed method could classify assistance motions for supporting standing-up with different foot positions. Participants were five young men and were asked to perform two motions with different two feet anteroposterior positions positions ("short step length" and "long step length"). The results showed that the proposed method was able to classify the two motions with 90% or higher accuracy for all participants. Therefore, we considered that the method could be used in the educational system.