Motion Capture System Using IMU Sensor with Mahony Filter

Nowadays, motion capture is applied in many fields, such as movies, video games, virtual reality, healthcare, sports, robotics, etc. Although systems using cameras and markers can be considered as the gold standard in the field of motion capture because of their accuracy and popularity, systems using IMUs offer more flexibility, much less setup time, and lower costs. This paper presents an IMU-based motion capture system. Full-body motions are real-time captured by 11 body-worn wireless sensors. A sensor fusion algorithm using Mahony filter is proposed to estimate the orientations for reconstructing human body motions by a 3D character model in Unity software. According to experimental results, the static error of the roll and pitch angle is less than 1 degree, and less than 2 degrees for the yaw angle.