Target localization and course change detection using bearing and bearing rate measurements

This paper addresses the problem of tracking a maneuvering target from passive measurements collected by a single observer. The target motion is characterized by several legs where in each leg the target moves with a constant speed and course. Course changes can be performed almost instantaneously. Furthermore, the observer takes either bearings-only or combined bearing and bearing rate measurements obtained by signals originating from the target. Here, the measurements are acquired by taking advantage of signal processing using a sensor array. In order to estimate the target state, a nonlinear Bayesian estimator, more precisely the marginalized particle filter, is applied in this work. Additionally, the times of course changes are detected employing a chi-square test. Finally, a comparison of the estimation results, obtained by Monte Carlo simulations, is performed between the cases of bearings-only measurements and combined bearing and bearing rate measurements. It is shown that a superior performance is achieved when using combined bearing and bearing rate measurements