Navigation and Probability Assessment for Successful AUV Docking Using USBL

Abstract

This paper addresses the navigation of an autonomous underwater vehicle (AUV) when approaching a docking station. The following array of underwater navigation sensors is considered available: ultra short baseline (USBL), Doppler velocity log (DVL) with bottom track, compass, and gyro. Based on this sensor array, we propose a modular and cascaded Kalman filter (KF) approach to estimate the vehicle's heading, azimuth and angle to the docking station, position, and velocity.We also propose to use the information from the KF covariance to compute, in real time, the probability of a successful docking. This method uses the navigation covariance and the characteristics of the docking station to predict the outcome of the docking procedure. The method enables the AUV to determine if a docking maneuver is feasible, hence improving the overall AUV autonomy level and allowing the AUV to assess whether corrective measures are required.