A flexible robust Student's t-based multimodel approach with maximum Versoria criterion

The performance of the state estimation for Gaussian space models can be degraded if are affected by non-Gaussian process and measurement noises with uncertain degree non-Gaussianity. In this paper, we propose a flexible robust Student's t-based multimodel approach. More specifically, degrees freedom parameter from t-distribution is assumed unknown modelled Markov chain values. order to capture more information t-distributions propagated through multiple models, establish model-based Versoria cost function in form weighted mixture rather than original form, maximize interact fuse models. Simulated results prove flexibility robustness proposed t-basedmultimodel approach when existence probability outliers uncertain.