Accuracy Analysis And Optimization Technology Of INS / GNSS Deep Coupling System For High Dynamic

With the development of military technology, the dynamic performance of the military aircraft is improving constantly. The requirements of navigation systems are increasing,single navigation system can not meet the demand any more. The deep integration of INS and GNSS is a frontier of integration navigation, which deeply exploiting and utilizing information from INS makes GNSS and INS deeply assist each other to implement a more precise and robust navigation system. Hence the INS/GNSS deep integration is a kind of effective method to solve the problem of high dynamic navigation. Furthermore, as the native Bei Dou Navigation Satellite System(BD) is being perfected gradually, the research on the INS/GNSS deep integration means a lot for the developing of the deep combination of INS and BD.Firstly, brief introductions to the navigation in high dynamics and the deeply integrated navigation system are given, including principles of INS and GNSS receiver, and a summary of system model widely used in current research of deep integration is presented. The Cubature Kalman Filter is choosen as the algrothim of deep integration by comparing with other fully developed nonlinear algrothim.Then the detail effects of high dynamic on some aspects of deep integration system like tracking, integration filter and loop feedback control are analyzed. According to the analysis, the thesis points out that the key to improve the accuracy of deep integration in high dynamic is to improve the perpormance of pre-filter.The study of pre-filter is carried out from model and algrothim. For the model of pre-filter, the thesis mainly rebuilds the measurement equation according to the characteristic of high dynamic GNSS signal, which would describe the measurements much more precisely and increase the precision of parameter estimation. For the pre-filter algorithm, the thesis mainly studies on utilizing a kind of Strong Tracking Filter based on Cubature Kalman Filter to defend outliers in measurement result and the inaccuracy of measurement equation. The algorithm utilizes Strong Tracking Filter as the frame of the algorithm to improve system robustness to high dynamic. Meanwhile, the method uses Cubature Kalman Filter to solve the non-linear filtering problem.Finally, the research achievements are verified on the simulation platform of deep integration system, which inclucing high dynamic trajectory generator, SINS simulator, high dynamic middle frequency signal generator, software GNSS receiver, filters, etc. The simulation results indicate that the optimization scheme proposed by this thesis can effectively improve the navigation accuracy in high dynamics.