New Approach for GPS/INS Integrated Navigation

The integration of position and velocity is commonly used for observation model in the GPS/INS integrated navigation system, but correlation between the system´s measurements of this observation model decrease the navigation accuracy. So the integration of pseudo-range was adopted for observation model to eliminate the influence of correlation. As the observation model is nonlinear, the unscented particle filter (UPF) was introduced for states estimation to eliminate the linearization error. The simulation was carried out under the condition of less than four GPS satellites data and the filtering results of UPF method were compared with the results of extended Kalman filter (EKF) method. The simulation results show that the estimated values with both UPF and EKF method are close to the real values but the estimation error of UPF algorithm is smaller than that of EKF method.