Terrain Aided Navigation for Long Range AUV operations at arctic latitudes

Abstract

This paper investigates the potential of use of Terrain Aided Navigation (TAN) methods to exploit the extended endurance capabilities of an enhanced 1500m depth rated variant of the Autosub Long Range (ALR), the ALR1500 Autonomous Underwater Vehicle (AUV), for operations under the Arctic ice. A simulator for the TAN system using a low resolution Arctic bathymetric map is developed to study the capability to reduce/bound the positioning error growth due to the Dead Reckoning (DR) drift over time. Bathymetric measurements are obtained using a low sample rated single beam echo sounder and a Jittered Bootsrap Particle Filter (JBPF) is used for the multi-sensor data fusion problem. Sensitivity analysis of JBPF has been performed for a range of values of number of particles, jittering variances and measurement frequencies and a discussion over the results is given. Results from the simulation demonstrate that JBPF is capable of providing a robust and accurate solution to the localization problem provided that the AUV follows a trajectory with a sufficiently variant terrain. Finally, a brief discussion regarding the future extension of this work is also provided.