Body Topology Recognition and Gait Detection Algorithms With Nine-Axial IMMU

Abstract

Generally, the positioning system is based on the wearable inertial and magnetic measurement unit (IMMU), but its positioning algorithm relies heavily on the wearing position, and there is no adaptive positioning algorithm for the wearable position. To solve this problem, this paper proposes a human body topology recognition (BTR) algorithm based on gait information, which can accurately and on real time identifies six typical wearing positions. The average recognition rate is 99.11%, and the real-time performance is controlled within 1.2 s. As a priori information, a gait detection algorithm based on the human BTR is extended. Experimental results show that the proposed algorithm can accurately classify the gaits of different wearing positions, achieving 99.39% average accuracy. This method realizes the flexibility, diversity, and portability of wearable IMMU. It can be widely used in attitude calculation, posture recognition, and dead reckoning of human or robot.