In the land vehicle GNSS/INS navigation system,the GNSS signal easily be out of lock because of being blocked or jammed,then the navigation accuracy would decreased.In order to solve this problem,virtual satellite algorithm was implemented to GNSS/INS integration.By adding virtual satellites,the virtual observations were constructed,and then the virtual observations and the real observations were able to be provided to carry on a solution for kalman filter.The result of the experiment indicated that the observability and navigation accuracy were able to be improved.

Considering the dynamic systems in the presence of additive white Gaussian noise,combining square-root squre-root unscented Kalman filtering and unscented Kalman filtering with additive noise,squre-root unscented Kalman filtering method without augmented state vector is proposed,and the according algorithm is deduced.The novel method has lower computational burden,better accuracy and better resistance to filter dispersion than traditional methods.The novel method is applied in GNSS multipath estimation,and it can estimate the parameters of the direct and multipath signal components efficiently and accurately.Simulant tests show the validity of the new methods proposed.

A novel method of differential coherent integration (DFC) algorithm combined with wavelet transform is proposed to acquire the weak global positioning system (GPS) signal parameters. The paper compared this proposed method with the traditional weak satellite signal acquisition algorithms to verify its feasibility. This method utilizes wavelet de-noising for the result of DFC to improve the Signal-to-Noise ratio (SNR) of acquisition process. Simulations demonstrate that the proposed method is better than DFC in obtaining larger acquisition output SNR.

On accounting of the divergence of general Kalman filter in SINS/GPS integrated navigation system.This paper introduces adaptive filter and H∞ filter.Analyzed their peculiarities.Last made simulation to prove the validity and usability of these two kinds of filter used for the SINS/GPS integrated navigation system.Thus,it is of significant in practice for filter design in integrated navigation system.

In order to solve the problems of the normal equation being ill-conditioned and observations containing outliers in GPS rapid positioning,a new robust estimate-regularized robust estimate based on the idea of robust M-estimation is proposed.Under this method can solve the robust estimation of ill-posed problems in geodesy.The computation results of GPS data show that the new estimator has the characteristics of resisting outliers,high precision and good stability.