Bearings-Only Tracking Based on Multiple Sensor Measurements and Generalized Particle Filtering

In this paper we address the problem of tracking by using bearings-only data obtained by more than one sensor. We apply the generalized particle filtering methodology which does not require any probabilistic assumptions, including prior probabilities and noise distributions in the state and observation equations. As a result, the proposed approach is much more robust in performance than standard particle filtering. We investigate the method when there is an exchange of information between the sensors. The advantage of the proposed method over standard particle filtering is illustrated through computer simulations.