Taylor Moment Expansion for Continuous-Discrete Gaussian Filtering

This article is concerned with Gaussian filtering in nonlinear continuous-discrete state-space models. We propose a novel Taylor moment expansion (TME) filter, which approximates the moments of stochastic differential equation temporal expansion. Differently from classical linearization or Itô-Taylor approaches, formed for functions directly and time variable, not by using on model. analyze theoretical properties, including positive definiteness covariance estimate stability TME filter. By numerical experiments, we demonstrate that proposed filter significantly outperforms state-of-the-art methods terms estimation accuracy stability.