Application of Robust Kalman Filtering to Integrated Navigation Based on Inertial Navigation System and Dead Reckoning

Aiming at solving the problem that the integrated navigation system which consists of inertial navigation system (INS) and dead reckoning (DR) has outliers and disturbance in measurement, as well as uncertainty in modeling, this paper proposes a new data processing method based on the technology of robust Kalman filtering (KF). The robust estimation is adopted to eliminate the outliers of the measurement of the sensors, thus the navigation information can be available for INS and DR respectively. Then for the INS/DR integrated navigation, the robust filtering is applied to resolve the modeling uncertainty. With this proposed method, experimental results show that the effect of the outliers in the measurement is eliminated, meanwhile, the robustness of the system is guaranteed, and therefore, the accuracy of integrated navigation is ensured.