Research On Integrated Navigation System And Filtering Algorithm

In 1970s,people begin to using integrated navigation system in spaceflight and seafaring. The study on the integrated navigation System based on Inertia Navigation System (INS) and Global Positioning System (GPS) is done in the paper.The GPS and INS are separately discussed in the paper. And systemic composition, orientation theory, error model and model calculation of GPS and INS are discussed. The development of the Kalman filtering is of great significance for the integrated navigation system. Normal kalman filtering have a high request in the application., especially for the inaccurate model parameters and noise covariance. It works worse and do not get optimized estimated values of systemic states, even do not converge.Many applied filter algorithm is occurred, including self-adaptive kalman filter. Some self-adaptive algorithms, Sage-Husa self-adaptive filtering algorithm, strong tracking Kalman filtering algorithm, maximum-likelihood kalman filtering and other improved self-adaptive algorithms, are proposed in this paper. Subsequently, simulations of normal Kalman filtering, extension Kalman filtering and self-filtering discussed in the paper are done in every different system. The working fields of each filtering and comparisons of their advantages and disadvantages is made in the paper. We design the Fuzzy Inference System, simultaneously builds the storeroom of the fuzzy rules and then combine it with the Kalman filtering to modify online the covariance matrix of system measurement noise. Simulations show this filtering algorithm can get a better result in system states optimization estimation. Simultaneously it has a superiority of adapting system noise variation and may greatly improve the precision of the navigation system.