Integrated particle Rauch-Tung-Striebel backward smoothing for single target tracking in clutter

Abstract

This paper presents a Rauch-Tung-Striebel backward smoothing (RTSBS) methodology applied to the integrated particle filter (IPF), which is an algorithm for single target tracking in clutter by incorporating the probability of target existence into the traditional particle filter as a track quality measure for false track discrimination (FTD). This integrated particle-Rauch-Tung-Striebel backward smoothing (IP-RTSBS) algorithm absorbs the forward filtering backward smoothing approach to smooth the trajectory state, which is then applied to the RTS smoothing methodology to obtain the smoothing propagation to update the probability of target existence. As a result, both the probability of target existence and trajectory estimation are improved. The smoothing benefits is validated in the simulations.