A Fuzzy-based Adaptive Unscented Kalman Filter for State Estimation of Three-dimensional Target Tracking

Although several state estimation methods have been presented for nonlinear systems, the challenges in implementing these estimation methods for unmanned aerial vehicles (UAVs) remain due to the presence of nonlinearities, uncertainties, and complex dynamics. In this paper, an adaptive approach of unscented Kalman filter (UKF) based on fuzzy logic control (FLC) termed as fuzzy adaptive UKF (FAUKF) is proposed to enhance the state estimation performance for three-dimensional bearing-only target tracking. The measurement noise covariance is adjusted based on an adaptation law to deal with noise uncertainties. The output of the Mamdani fuzzy inference system (FIS) is used as a tuning factor in the adaptation law. The performance of the proposed method is compared with that of conventional UKF by root mean square error (RMSE) of states and the mean and standard deviation of these errors through the simulation of 500 Monte Carlo runs. The simulation results show that the FAUKF algorithm for UAVs does a much better job of estimating the state than conventional UKF and AUKF.