Exploration of adaptive filters for target tracking in the presence of model uncertainty

This paper presents an investigation of Target Motion Analysis (TMA) algorithms that are designed to cope with some model uncertainty. In particular, adaptive algorithms are designed to deal with unknown noise variance. These adaptive algorithms are multiple model based techniques that are capable of tuning into the true parameter while estimating the target state. The algorithms considered are a) Static Multiple Model (SMM) Estimator, b) Generalised Pseudo Bayes (GPB) methods, and c) Interacting Multiple Model (IMM) based tracker. Simulation results verify the potential use of such algorithms.