Fault Detection and Isolation for GPS RAIM Based on Genetic Resampling Particle Filter Approach

An integrity monitoring system is an inseparable part of global positioning system (GPS). According to the measurement noise feature of GPS receiver and the degeneracy phenomenon and alleviating the sample impoverishment problem in particle filter (PF). An approach to fault detection and isolation (FDI) for GPS receiver autonomous integrity monitoring (RAIM) based on genetic resampling particle filter is proposed. The genetic algorithm (GA) is melted into the re-sampling process of the basic particle filter to solve the particles degeneracy and impoverishment problem. A main GA-aided particle filter (GPF) is used to process all the measurements to produce the optimal state estimate, several auxiliary GPFs are used to process subsets of measurements to produce the state estimate as detection references. By setting up the log-likelihood ratio (LLR) test to check the consistency of test statistics. The particles in GPF are assorted by weights, in order to reduce the computation complexity of the algorithm, only the lower weight particles participate in genetic operations. By collecting the GPS data from the GPS reciver, the feasibility and effectiveness of the RAIM approach is verified, and comparing with extended Kalman filter (EKF) and PF algorithm. The results show that the approach in the case of non-Gaussian measurement noise can estimate the state accurately, also can successfully detect fault satellite, therefore, improve the reliability of GPS positioning.