基于鲁棒扩展卡尔曼粒子滤波的RAIM算法

Since the problems of particle degeneracy and sample impoverishment exist commonly in the particle filter used in the receiver autonomous integrity monitoring (RAIM) algorithm, a RAIM algorithm based on robust extended Kalman particle filter (REKPF) is proposed. In this method, the extended Kalman filter is used to calculate the proposed density function, so that the sampling distribution of re-sampling is more accurate. Meanwhile, in order to reduce the influence of pseudo-range bias on the filter estimation, the Kalman gain matrix is corrected by robust estimation. Based on the real global position system data, the statistic of consistency test of satellite fault detection is established, and then the cumulative logarithmic likelihood ratio of each state is compared to detect the faulty satellite. Simulation results demonstrate that, when there is a pseudo-range bias on a satellite, the RAIM algorithm based on REKPF can diagnose the faulty satellite effectively, shorten the alarm delay time, and improve the position accuracy, thus the performance is better.