TY - GEN
T1 - Adaptive Maximum Correntropy Unscented Kalman Filter for Aero-Engine State Estimation
AU - Wang, Guangfeng
AU - Gou, Linfeng
AU - Huang, Yingzhi
AU - Chen, Yingxue
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - 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.
AB - 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.
KW - adaptive
KW - maximum correntropy criterion
KW - non-Gaussian noise
KW - unscented kalman filter
UR - http://www.scopus.com/inward/record.url?scp=85186744314&partnerID=8YFLogxK
U2 - 10.1109/ICMAE59650.2023.10424042
DO - 10.1109/ICMAE59650.2023.10424042
M3 - 会议稿件
AN - SCOPUS:85186744314
T3 - 2023 14th International Conference on Mechanical and Aerospace Engineering, ICMAE 2023
SP - 230
EP - 236
BT - 2023 14th International Conference on Mechanical and Aerospace Engineering, ICMAE 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Conference on Mechanical and Aerospace Engineering, ICMAE 2023
Y2 - 18 July 2023 through 21 July 2023
ER -