IMMPDA filter via perception net

Abstract

This paper presents an improved performance of the IMMPDA (interacting multiple model with probabilistic data association) filter with the detecting capability of a maneuvering. In order to detect the maneuvering of a target, an error monitoring and recovery method of perception net is applied to the IMMPDA filter. This method is derived as an extension of the previous results given by Choi et al. (1999) on tracking for a maneuvering target without clutter. The effectiveness of the proposed method is validated through simulations by comparing it with the conventional IMMPDA filter.