An adaptive fast quaternion-based human motion tracking algorithm with Inertial/Magnetic technology

Real-Time human motion tracking with Inertial/Magnetic sensors is a key technology in human-computer interaction applications. A computationally efficient and robust algorithm for estimating orientation is critical. This paper gives a quaternion-based orientation tracking algorithm for human motion tracking with a triaxial angular rate sensor, a triaxial accelerometer, and a triaxial magnetometer. The algorithm utilizes the Gauss-Newton method for parameter optimization in conjunction with our adaptive filter. Based on the use of the Gauss-Newton method, particularly the reduced-order implementation, this paper presents a method to avoid computing matrix inverse compared to the conventional G-N method to decrease the computing time. In addition, considering the effects of ferromagnetic magnetic disturbances and body motions acceleration, the paper proposes an adaptive filter to accommodate the disturbance situation. Real-time implementation and testing results of the quaternion-based algorithm are presented. Experimental results validate the algorithm and show the robustness and efficiency for human motion tracking.