A fixed-lag smoothing algorithm for Markovian switching systems

A suboptimal approach to the fixed-lag smoothing problem for Markovian switching systems is presented. We augmented the system state and system model probability, so the fixed-lag smoothing problem is transformed to the filtering problem of new augmented system. A new fixed-lag smoothing algorithm is developed by applying the basic structure of IMM, which is superior in MME methods, to the augmented systems. The algorithm is illustrated via two typical target tracking simulation examples where a significant improvement over the Chen's algorithm, which just arguments the state vector, is achieved at the cost of a time delay.