Overview of the multisensor IMMPDA filter with an amplitude feature for tracking maneuvering target in cluttered environment

Abstract

This paper gives an overview of the multisensor IMMPDA filter with an amplitude feature for tracking highly maneuvering target in clutter. Interactive multiple model (IMM) filter was already described in a number of excellent papers and is proven to be the most cost-effective solution to the maneuvering target tracking problem. Probabilistic data association (PDA) approach is an extension to this filter for tracking target in cluttered environment. Data fusion from radar and more accurate infrared sensor gives further improvement in tracker robustness during maneuver. Yet another method for improving tracking is augmentation of target state with amplitude information from radar. Results from computer simulation of the combined radar and infrared sensor are discussed, with and without amplitude feature and with various models used in IMM filter. There is motivation to explore further tracking enhancement by augmenting target state vector with both the amplitude and the frequency information from radar and to investigate in filter robustness with more realistic nonlinear nongaussian ground target behavior.