Abstract
This paper proposes the measurement method of 3D orientation using inertial sensors. The method estimates the SDorientation of the local coordination by inertial sensors attached to the two link body segment. The sensor fusion that is established in this study estimate 3D orientation compensated drift error of gyro sensor output by translational and gravity acceleration of accelerometer output. The nonlinear state equation in the sensor fusion can estimate 3D orientation (Roll-Pitch-Yaw angle) by applying to extended Kalman filter. In order to indicate the performance of the proposed method, we have conducted the experiment to confirm precision of the proposed method using the system installed the rotary encoder. As the result, the proposed methods have estimated 3D orientation compensated drift error of gyro sensor output. The result of method by unscented Kalman filter has indicated higher precision than the result of method by extended Kalman filter. The proposed method can be used to the orientation measurement in various motions.
