Non Gaussian State Estimation and the Maximum Correntropy Approach

Abstract

This monograph aims to present the recent advances in state estimation, in terms of relaxing the conventional assumption that probability densities remain Gaussian. The book explains how MCC is integrated into the conventional Bayesian estimation framework and their implementation to real-life problems. Features: Reviews well-established non-Gaussian estimation methods including applications of techniques  Covers relaxation of gaussian assumption                                                                                                        Discusses challenges in formulating non-liner non-Gaussian estimation framework                 Illustrates the applicability of the algorithms mentioned to real-life problems                                               Explores derivation of non-linear non-Gaussian estimation framework based on maximum correntropy criterion                                                                                                                                This book is aimed at researchers and graduate students in electrical engineering, robotics, and dynamic systems.