Hidden Markov Model Based Variable Structured Multiple Model Algorithm

Abstract

In this study, a novel methodology is proposed for more accurate target tracking. The proposed method is basically built on the merging of interacting multiple model (IMM) structures using Hidden Markov Model (HMM). Thus, more models are used than the ordinary IMM algorithm, but more accurate state vectors are estimated by selecting the most likely ones. The proposed algorithm is compared with the variable structure IMM (VSIMM) algorithm, which is the most similar methodology in the literature, in MATLAB environment and the results are presented.