Federated filtering algorithm based on fuzzy adaptive UKF for marine SINS/GPS/DVL integrated system

Abstract

For marine SINS/GPS/DVL integrated navigation system, a new federated filtering algorithm based on fuzzy adaptive unscented Kalman filter is presented in order to satisfy the precision and fault-tolerance demands. The algorithm uses unscented Kalman filter (UKF) to estimate the nonlinear system states, as well as using fuzzy inference system (FIS) to amend measurement noise adaptively. The simulation in marine SINS/GPS/DVL integrated system illustrated that, compared with the traditional federated Kalman filtering algorithm, the algorithm in this paper can improve the navigation precision and fault-tolerant capability effectively, the velocity-error decreased from 0.14m/s to 0.06m/s.