A novel self-adapting filter based navigation algorithm for autonomous underwater vehicles

This paper presents a novel approach to the design of globally asymptotically stable position and assessment of an USBL-aided integrated navigation based on Condition-adaptive gain Extended Kalman Filter (CAEKF) for the deep water Autonomous Underwater Vehicles (AUVs) subject to uncertainties (e.g., loss of USBL signal, irregular and gross positioning error). Due to the influence of underwater observation conditions, positioning gross error will often appear when exploiting USBL to assist AUV navigation in ocean exploration. Aiming at this kind of problem a method of adding the conditional constraints and confidence assessment to EKF was put forward to filter the positioning value of USBL, and which can make the filtering result more robust and smooth. In addition, in order to reduce positioning error for the deep water vehicle online, an integrated navigation system is constructed by adding the acoustic navigation. Finally, the long voyage of the sea-trials data acquired in suitable sea trials performed in the South China Sea verifying the robustness and practicability of the proposed methodology, a very effective trade-off between accuracy and computational load has been achieved, and which demonstrated that the proposed algorithm outperforms standard navigation algorithms and other classical filtering approaches.