GPS/SINS Navigation Data Fusion Using Quaternion Model and Unscented Kalman Filter

摘要

A generic Strapdown Inertial Navigation System (SINS) error propagation model is presented. Quaternion is used for relative attitude representation and attitude error interpretation. The resultant error propagation model is nonlinear and does not rely on small misalignment angles assumption. Based on this model, SINS algorithm is developed for en-route alignment Global Positioning System (GPS) is used as external sensor to confine the SINS error. Unscented Kalman Filter (UKF) is used for the non linear data fusion. Numerical simulation reveals validity of the model and superior performance of UKF.