A novel drift compensating method for orientation measurement system in VR applications

Accurate and fast estimation of a target's orientation with inertial/magnetic sensors is widely applied in many Virtual Reality applications. Due to the abrupt movement of targets, the accuracy of the motion estimate degrades. In this paper, we propose a quaternion-based Kalman filter for a multi-sensor system which utilizes a low-cost tri-axis accelerometer, a tri-axis gyroscope and a tri-axis magnetometer. The proposed Kalman filter combines a novel error model in its measurement update equation which makes it possible to separate the acceleration error from the gravitational acceleration, and achieve an accurate orientation estimate. Due to the computational efficiency of the proposed algorithm, the system is suitable for human motion tracking applications in real-time. Results of simulations and experiments prove the efficiency of the proposed algorithm.