Preview

PDF 129542 - A mode...pdf | Download (474kB) | Preview

Abstract

Localization and navigation systems with a lower acoustic signature is a requisite forautonomous underwater vehicles (AUVs) carrying out surveillance operations, which is achallenging task due to limited availability of sensors. Utilizing inertial navigation systems(INSs) is one of the options, although it still presents a certain degree of position drift, even witha high-grade INS. This study investigates the capability of a model-aided inertial navigationsystem (MA-INS) with limited activation of Doppler velocity log (DVL) acoustic sensorto localize AUVs. Employing a Kalman filter (KF) data fusion algorithm, the accelerationmeasurements from the INS and the vehicle velocities predicted by a motion response predictingmathematical model were combined with intermittent DVL measurements. The acousticsignature was reduced while maintaining a reliable localization accuracy. These findings showthat localization of surveillance AUVs can be conducted by the MA-INS algorithm with aminimized DVL usage. As the DVL operating time could be controlled by the DVL applicationschemes, the algorithm could be optimized depending on the planned routes of AUV duringthe field trial. However, considerable investigation may be needed to obtain highly accuratelocalization solution using non-acoustic sensors. For future work, it is recommended toimpeccably restrain the use of DVL for the localization, in order to secure the concealment ofAUV from detection.

Item Type:	Article
Authors/Creators:	Choi, B and Randeni Pathiranachchilage, SAT and Fan, S and Forest, A
Keywords:	autonomous underwater vehicle, dynamic modeling, control and estimation, localization algorithm, mathematical model aided inertial navigation system, surveillance AUV
Journal or Publication Title:	Journal of Ocean Technology
Publisher:	Memorial University of Newfoundland, Fisheries and Marine Institute
ISSN:	1718-3200
Copyright Information:	Copyright 2018 Journal of Ocean Technology
Item Statistics:	View statistics for this item

Actions (login required)

Item Control Page