An information-theoretic approach to underwater magnetic dipole localization

Algorithms for underwater magnetic anomaly localization are presented for autonomous underwater vehicle applications. Two approaches are considered. The first, based on information theoretic methods, utilizes an extended information filter to localize a single dipole. This approach enables an autonomous underwater vehicle to choose trajectories that locally maximize information gain and to speed up localization. The second approach utilizes an occupancy grid framework and the theory of evidence, and it is suitable for multiple dipoles.