Research on Algorithm of Robust Filtering in SINS/GPS/CNS Integrated Navigation System

Integrated navigation system is difficult to model accurately and statistical properties of noise are difficult to obtain. These increasingly prominent problems affect the stability of Kalman filter. This article applies. To improve the rubost of SINS / GPS / CNS integrated navigation system,the article designed the algorithm in the case of changing parameters of the system and compared the filter results of H∞filter and Kalman filter. The results show that under the condition of model change,the precision of H∞filtering algorithm is almost the same. Therefore,in the case of noise statistics characteristics and model parameters is difficult to determine,H∞filtering has better adaptability and improves the performance of the navigation system.