Fast Monte Carlo localization of AUV using acoustic range measurement

This paper presents a novel online nonlinear Monte Carlo algorithm for multi-sensor autonomous underwater vehicle (AUV) navigation. The approach integrates the global constraints of range to, and GPS position of, multiple surface vehicles communicated via acoustic modems and relative pose constraints arising from observations of multiple beacon boats. The proposed method can be used to more accurately navigate the AUV, to extend mission duration, and to avoid surfacing for GPS fixes. The Monte Carlo method is used for the estimation of the AUV pose. Although it is also desirable to estimate the range measurement of the surface vehicles using a particle filter (PF), implementing a PF for each beacon onboard the AUV is computationally expensive. Thus for the range estimation, an extended Kalman filter (EKF) is proposed for each beacon. We discuss why our approach is more computationally efficient and suitable for use on underwater vehicles. Simulation results are provided for AUV navigation using multiple autonomous surface vehicles (ASVs) in an ocean environment. During these simulations the proposed algorithm runs online on-board the AUV. In-water validation is currently in progress.