RAIM Algorithm Based on Robust Extended Kalman Particle Filter and Smoothed Residual

Abstract

With the rapid development of Global Navigation Satellite System (GNSS), receiver autonomous integrity monitoring (RAIM) has become an essential part of integrity monitoring of navigation satellite system. Conventional RAIM algorithm requires the observation noise obeying Gaussian distribution, but in some conditions it obeys non-Gaussian distribution. Particle filter is applicable to nonlinear and non-Gaussian system, thus RAIM algorithm performance is improved under non-Gaussian noise with particle filter method. However, particle degeneration interferes the performance of particle filter. In this paper, robust extended Kalman particle filter is proposed and applied to receiver autonomous integrity monitoring. The importance density function of particle filter is calculated by robust extended Kalman filter in order to improve the accuracy of state estimation when pseudo-range bias exists, and the particle degeneration is restrained. On this basis, the smoothed residual test statistics is set up for satellite fault detection and isolation. The simulation results show that RAIM algorithm based on robust extended Kalman particle filter and smoothed residual can well detect and isolate the faulty satellite under the condition of non-Gaussian observation noise. Compared to the RAIM algorithm based on particle filter, the new RAIM algorithm has a better performance on fault detection, and its position accuracy is improved.