Research on Attitude Updating Algorithm Based on Dual Quaternion for Attitude and Heading Reference System

The parameter model for attitude updating in the field of micro-strapdown attitude and heading reference system (AHRS) was established based on the dual quaternion (DQ) algebra. Differing from the quaternion (Q) solution adopted to express the pure rotation transition of rigid body, DQ scheme described the coordinates transition between body frame and geographical frame by fusing the measurements of rotation and translation process, computed the optimized attitude updating matrix by constituting the least squares cost function as well. Moreover, the Kalman filter functions were designed, which then defined the real-time attitude and heading estimates. Simulation indicates that the mean and root mean square of attitude quaternion error based on DQ technique are both below 0.002. The modeling of Kalman filter accords with the requirements for practical applications, the pitch, roll, and heading angles error never exceed 0.1°, 0.1°, 0.2° respectively, even if the accuracy of the sensors is modest.