A novel adaptive maximum correntropy cubature Kalman filter based on multiple fading factors
Peng Gu,Zhongliang Jing,Liangbin Wu
DOI: https://doi.org/10.1177/01423312241229039
IF: 2.146
2024-02-16
Transactions of the Institute of Measurement and Control
Abstract:Transactions of the Institute of Measurement and Control, Ahead of Print. In this paper, an adaptive maximum correntropy cubature Kalman filter based on multiple fading factors (MAMCKF) is proposed to address the problem of inaccurate process noise covariance and unknown measurement noise covariance together with outliers in target tracking. Although there are many adaptive filters and robust filters have been proposed to handle unknown measurement noise covariance or measurement outliers, most filters cannot deal with both unknown noise covariance and outliers simultaneously. In this article, we propose an adaptive and robust cubature Kalman filter. The modified measurement noise covariance matrix (MNCM) and innovation covariance matrix are used to construct multiple fading factors for correcting the prediction error covariance matrix (PECM), which can achieve adaptability. Then, the maximum correntropy criterion (MCC) is introduced to suppress outliers, which further enhances the robustness. Compared with the existing approaches, the proposed approach improves the performance by at least 5% in unknown time-varying noise, unknown time-varying heavy-tailed noise, and non-Gaussian heavy-tailed noise scenarios. The simulation results show that the proposed approach can effectively suppress inaccurate process noise covariance and unknown time-varying measurement noise together with outliers. Compared with the existing filtering approaches, the proposed approach exhibits both adaptability and robustness.
automation & control systems,instruments & instrumentation