Sidescan sonar aided inertial drift compensation in autonomous underwater vehicles

We introduce our first work on terrain-based navigation of autonomous underwater vehicles using sidescan sonars and sidescan sonar maps. It consists in estimating the state of the vehicle from its inertial proprioceptions and sidescan exteroceptions using a particle filter. The novelty is the use of sidescan acoustic perceptions and sidescan sonar maps instead of single-beam bottom-looking acoustic perceptions and bathymetric maps. Also, the approach is not landmark-based. Our first simulations, although conducted under simplifying hypotheses, show that the approach is valid and may be applied to more complex situations.