A new approach for planar tracking in a nongaussian setting

This paper describes a new efficient approach to the conventional nonlinear tracking problem in a nongaussian setting that consists in the transformation of the nonlinear output measurement function in a linear form by the definition of a virtual measurement process. Such a procedure leads to the use of an efficient filter capable to take into account the nongaussanity of the transformed measurement noise process. This key feature is also exploited to consider and suitably manage a nongaussian and more realistic motion behaviour of the target object. Compared with the traditional approaches (e.g., extended Kalman filter (EKF) and unscented Kalman filter (UKF)) used in passive localization, the proposed method has potential advantages in robustness, convergence speed, and tracking accuracy.