A Novel Robust Interacting Multiple Model Algorithm for Maneuvering Target Tracking

In this paper, the state estimation problem for discrete-time jump Markov systems is considered. A minimax filtering technique, interacting multiple model algorithm based on game theory, is developed for discrete-time stochastic systems. Filter performance improvement in presence of model uncertainties, measurement noise, and unknown steering command of the maneuvering target is illustrated. It is shown that the technique presented in this paper has a better performance in comparison with the traditional Kalman filter with minimum estimation error criterion for the case of worst possible steering command of target. In particular, simulation results illustrate the improved performance of the proposed filter compared to Interacting Multiple Model (IMM), diagonal-matrix-weight IMM (DIMM), and IMM based on H. (IMMH) filters.