A study on error reduction of joint torque estimation using wearable sensor system

Abstract

This study aims to error reduction in estimation of joint torque using wearable sensor system. The wearable sensor system consists of wearable motion sensor installing the inertial and magnetic field sensors and wearable force plate installing several compact force sensors and inertial and magnetic field sensors. The wearable motion sensor outputs the angular velocity, the acceleration, and magnetic field, and the wearable force plate outputs the 6-axis force components and the motion sensor output. We conducted the measurement experiment of gait analysis using the wearable sensor system and the optimal motion capture system. We established the compensation method of error reduction in the effect by attaching on skin and clothes. This method was assumed the effect of error elements by using the spring and damping elements. We compared the posture information of knee joint in the results of without compensation and with compensation. The analytical results of the Roll angle indicated the effectiveness of this method.