Compensation of time lag between actual and virtual spaces by multi-sensor integration

Unconstrained measurement of human head motion is essential for HMD (Head Mounted Display) to be really interactive. Polhemus sensors developed for that purpose have deficiencies of critical latency and low sampling rate. This paper proposes a method which compensates the latency and raises the effective sampling rate through the integration of Polhemus and gyro sensors, For this purpose, the authors derived a physically consistent system model of human rotational motion described by quaternion and angular velocity. This integration consists of the optimal prediction based on simultaneous measurement of both sensors and the extrapolation of the estimate based on measurement of the gyro sensor. This paper also reports the implementation of the proposed method