A robust fixed-interval smoother for nonlinear systems with non-stationary heavy-tailed state and measurement noises

We propose a robust fixed-interval smoother for nonlinear systems with non-stationary heavy-tailed state and measurement noises, in which the state and measurement noises are modelled as Gaussian-Student's t mixture distributions. The variational Bayesian technique is utilized to deduce the smoother approximately. The standard cubature Kalman smoother (CKS) and the robust Gaussian approximate smoother (RGAS) with fixed scale matrices and dof parameters are two particular cases of the proposed smoother. Numerical simulation and target tracking example show the merits of the proposed smoother. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

A robust fixed-interval smoother for nonlinear systems with heavy-tailed state and measurement noises is proposed, where noise distributions are modelled as Gaussian-Student's t mixtures. The merits of this smoother are demonstrated through numerical simulation and target tracking examples.