Design and Implementation of Low-Cost Inertial Sensor-Based Human Motion Capture System

Motion capture system can transform motion information of human body into digital data for analysis and processing in computer or cyberspace. It is an important medium to connect physical space and cyberspace. Aiming at the problems of high cost and high power consumption of existing inertial motion capture systems, this paper proposes a wearable wireless motion capture system based on nine-axis inertial sensor and Bluetooth Low Energy technology. The system can capture the motion trajectory of human body and realize real-time human motion display and data storage in computer. An explicit complementary filtering algorithm based on proportional integral model is used in the process of character motion update. Motion capture experiments are carried out on a human body and results show that the system has sound dynamic performance with the error less than 2 degrees and can achieve efficient human motion capture. The sampling frequency of the system reaches 125Hz and the working time is up to 10 hours. The system can serve as the basis for developing affordable human motion capture solutions that require long-term testing without precise kinematic analysis.