Robust fixed-lag smoothing under model perturbations

A robust fixed-lag smoothing approach is proposed in the case there a mismatch between nominal model and actual model. The resulting smoother characterized by dynamic game two players: one player selects least favorable prescribed ambiguity set, while other minimizing error with respect to We propose an efficient implementation of smoother. Moreover, we characterize corresponding over finite time horizon. Finally, test examples. first regards target tracking problem, second parameter estimation problem.