A Robust Localization Method for Unmanned Surface Vehicle (USV) Navigation Using Fuzzy Adaptive Kalman Filtering

Recently, multi-sensor navigation has emerged as a viable approach in autonomous vehicles' development. Kalman filtering has been widely applied in multi-sensor data fusion, and researchers are trialing variants of the Kalman Filter (KF) to improve the operational robustness of vehicles in a range of environments under varying dynamic constraints. This paper proposes a novel sensor data fusion algorithm employing an Unscented Kalman Filter (UKF) for the autonomous navigation of an Unmanned Surface Vehicle (USV). Since the navigational sensors on-board the USV are subject to operational uncertainties caused by equipment limitations and environmental disturbances, an improved UKF algorithm with the capability of adaptive estimation, namely fuzzy adaptive UKF data fusion algorithm, has been proposed to obtain reliable navigational information. The conventional UKF is capable of fusing a number of raw sensor measurements and generating relatively accurate estimations with proper a priori knowledge of system noise. To deal with systems that lack such information, a fuzzy adaptive estimation method is introduced to enhance the performance of the conventional UKF, making the algorithm capable of verifying and correcting the associated sensor noise in real time. The proposed fuzzy adaptive UKF data fusion algorithm has been tested and evaluated in different simulations modeled using practical maritime environments and the results are compared with the conventional UKF. The sensor measurements taken from a practical USV trial have also been applied to the proposed algorithm for further validation.