Of .gps / Ins Integrated Navigation System Of Nonlinear Filtering Method

The advangtage of GPS/INS integrated navigation is discussed and the common filtering methods in GPS/INS integrated navigation are introduced in this paper. The algorithm principle and flow of the these algorithms are described detailedly. In order to make use of GPS navigation data, choosing the proper observation model and UPF(Unscented Kalman Filter) algorithm is proposed for filtering in GPS/INS integrated navigation system.To validate the availability of UPF algorithm in the GPS/INS navigation system,a large amount of simulations were carried out.According to comparing the filtering results of EKF(Extended Kalman Filter),UKF(Unscented Particle Filter), PF(Particle Filter) and UPF in the GPS/INS navigation system,UPF is proved to have better filtering results than the other filtering methods.This thesis analyzed the advantage and disadvantage of each GPS and INS, and then the superiority of GPS/INS integrated navigation is discussed. On the basis of it, by discussing the state equation and observation equation detailedly, the non-linear observation model integrated with pseudo-range and pseudo-range rate is selected. By optimizing the filtering algorithm, accurate navigation is achieved in case of lacking 4 GPS satellites.As to compare,the present filtering methods in GPS/INS integrated navigation system were analyzed, such as EKF, UKF and PF algorithm.And the simulation results of applying these methods in GPS/INS integrated navigation were analyzed. Based on this, UPF algorithm that is suitable for non-Gaussian and nonlinear systems is proposed for filtering in the navigation system.By carrying out a large number of simulations,UPF method is proved available in GPS/INS integrated navigation system. According to analysis and comparison of the simulation results,it shows that UPF method is an available filtering algorithm for GPS/INS integrated navigation system and UPF has a better filtering results than the other algorithms.