A Gaussian Filter for Nonlinear Systems with Correlated Noises and Unknown Sensor Measurement Loss

This paper introduces a new Gaussian filter for nonlinear systems plagued by correlated noises and unknown measurement loss. It tackles the issue of correlated noises by employing a Gaussian approximation recursive filter (GASF). To approximate the nonlinear aspects, it utilizes the Unscented Transformation (UT). Furthermore, it estimates measurement loss using the Maximum a Posteriori (MAP) criterion. Through simulations, this paper demonstrates the effectiveness of the proposed algorithm under measurement loss and correlated noises.