Adaptive Maximum Correntropy Unscented Kalman Filter for Aero-Engine State Estimation

Abstract

In this paper, we investigate the problem of state estimation for a class of non-linear systems with non-Gaussian measurement noise. Based on the maximum correntropy criterion (MCC), an adaptive maximum correntropy unscented kalman filter (AMCUKF) is derived by introducing a weighted combined cost function and an adaptive kernel function bandwidth. The filter solves the numerical problem of the existing maximum correntropy unscented Kalman filter (MCUKF) when the measured value contains large outliers and the problem of performance degradation caused by improper kernel bandwidth selection. Finally, taking the aero-engine state estimation problem as an example, the filtering performance of different filters is compared, which shows that the filter proposed in this paper has advantages in dealing with non-linear and nonGaussian systems.