Parameter identification in switching multiple model estimation and adaptive interacting multiple model estimator (Q2766016)

Switching Multiple Model Estimation (SMME) is widely used in many cases, such as target tracking and fault detection and its isolation, in which structures and parameters are uncertain and/or changed. Any online method for SMME should include identification of parameters, computation of Markov transition probabilities and the variances of the model-conditional process noise. Combining the parameter estimation with the Interacting Multiple Model (IMM), which is one of the most cost-effective estimations in SMME, the authors present an adaptive IMM and give a numerical simulation example for tracking a maneuvering target, which appears to be much more accurate than the usual IMM.