The shifted Rayleigh filter for bearings only tracking

The recently introduced shifted Rayleigh filter is a moment matching algorithm that exploits the essential structure of the nonlinearities present in bearings-only tracking. The algorithm can befitted to a wide range of scenarios and places no restrictions on model dimensionality. The key feature is that it generates the exact updated conditional distribution of target motion, given a normal approximation to the prior. In this paper, two versions of the algorithm are applied to the problem of tracking a moving object from multiple, independently drifting sonobuoys that supply noisy bearings only measurements, corrupted by clutter. A separate moving sensor provides noisy bearings only measurements of sonobuoy motion. The shifted Rayleigh filter adapts well to this scenario. Simulations indicate that it give good estimates, even in adverse circumstances when the clutter probability is 67% and the standard deviation of sensor noise is 16°.