Tight-integration adaptive filtering method of resisting to outliers of global positioning system

The invention provides a tight-integration adaptive filtering method of resisting to outliers of a global positioning system, and relates to an adaptive filtering method suitable for resisting to the outliers of the GPS and measuring noise time variation of a GPS/inertia tightly integrated navigation system. According to the tight-integration adaptive filtering method of resisting to the outliers of the GPS, a residual adaptive filtering method is combined with fuzzy logic control, the residual sequence of a filter is monitored through a fuzzy logic controller, and therefore measurement noise intensity is estimated online to adaptively adjust measurement noise matrix parameter values of the filter; thus, filtering accuracy is improved, and filter divergence is prevented. Meanwhile, on the basis of the filter residual analytical method, outlier detection of each satellite of the GPS in the tightly integrated navigation system is achieved in an assisting mode, and compensation for the outliers is achieved with