A new smoothing particle filter for tracking a maneuvering target

A new smoothing particle filter is presented for applications in maneuvering target-tracking problems. In practical, these stochastic dynamic systems are usually nonlinear and incompletely observed, and the main difficulty of maneuvering target-tracking problem lies on the fact that the maneuverability at every step is of highly uncertainties. We propose here a new smoothing particle filter algorithm, which combines the particle filter, which tackles the non-linear and non-Gaussian peculiarities of the problem, and smoothing of the PDF of system modes, which settles the maneuverability of the target. In a simulation comparison with the auxiliary particle filters, we show that our approach has superiority and yields performance improvements when tracking a maneuvering target.