The Research of Motion Capture Technology Based on Inertial Measurement

Abstract

The motion capture technologies have been widely used in the following areas: human movement science, human-computer interaction and control, medical analysis, film, game production, and etc. This paper presented a motion capture system based on inertial sensors. The system is mainly composed of the inertial sensor unit and real-time monitoring unit on PC. Meanwhile, Sensor units exchange data with PC through BSN (body sensor network). One sensor measurement unit contains a three-axis gyroscope, a three-axis accelerometer, a three-axis magnetometer and a network communication unit. The inertial sensor units are installed on the different parts of body. In the system, the inertial sensor data are collected, and then adaptive Kalman Filter algorithm is used to estimate the body gesture. The Experiments show that this way have some advantages such as: high precision, dynamic, and almost no drift. Furthermore, the data can be changed smoothly.