Toward human motion sensing: Design and performance evaluation of a minimized wearable platform using inertial and TOA sensors

Abstract

Wearable sensors based applications have been widely used in various applications, such as health care, security monitoring and human-computer interaction. Despite remarkable research efforts and many encouraging advances in the past decade, accurate recognition of the human motions is still a challenging task. The existing studies mostly face the problems of sensor drift errors and the assumption of independent body part movement. In this study, we take the geometrical relation among joints into consideration and a corresponding platform is designed to obtain potential parameters and perform human motion capturing. CRLB is derived to evaluate the performance of proposed platform. The lower bound of motion sensing accuracy shows apparent reduction with applying our method. The sensing accuracy has an obvious promotion in the experiment results.