Maximum correntropy sparse Gauss–Hermite quadrature filter and its application in tracking ballistic missile

Qin, Wutao; Wang, Xiaogang; Cui, Naigang

doi:10.1049/iet-rsn.2016.0594

Cited by 41 publications

(23 citation statements)

References 25 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dynamic GT model is selected as the state model (Li and Jilkov, 2010; Qin et al, 2017). In the dynamic GT model, the total acceleration in Earth Centered Initial Coordinate System (ECI-CS) is given as…”

Section: Target Tracking Models In Boost Phasementioning

confidence: 99%

Robust filtering with randomly delayed measurements and its application to ballistic target tracking in boost phase

Qin

Wang

Cui

2019

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

Motivated by the performance degradation of High-degree Cubature Kalman Filtering (HCKF) in coping with randomly delayed measurements in non-Gaussian system, a novel robust filtering named as Randomly Delayed High-degree Cubature Huber-based Filtering (RD-HCHF) is proposed in this paper. At first, the system model is re-written by the Bernoulli random variables to describe the randomly delayed measurements. Then, the Randomly Delayed HCKF (RD-HCKF) is derived based on the rewritten system model and 5th-degree spherical-radial cubature (SRC) rule. In order to enhance the robustness of the filter in glint noise case, the measurement update of RD-HCKF is modified by the Huber technique, which is essentially an M-estimator. Therefore, the proposed RD-HCHF is not only robust to the randomly delayed measurements, but also robust to the glint noise. In addition, the RD-HCHF is applied to the ballistic target tracking in boost phase, and the Gravity-Turn (GT) model is taken as the target model. Finally, the simulation is conducted and the tracking performance of RD-HCHF is compared with that of HCKF, RD-HCKF and High-degree Cubature Huber-based Filtering (HCHF). The results clearly confirm the superiority of the RD-HCHF.

show abstract

Section: Target Tracking Models In Boost Phasementioning

confidence: 99%

Robust filtering with randomly delayed measurements and its application to ballistic target tracking in boost phase

Qin

Wang

Cui

2019

Transactions of the Institute of Measurement and Control

Self Cite

View full text Add to dashboard Cite

show abstract

“…The similar situation applies to the improved algorithms based on EKF, including the modified polar coordinate EKF [4] and the range‐parameterised EKF [5]. Based on the idea of selecting a list of deterministic points to approximate the probability density distribution (PDD), non‐linear filtering algorithms such as the unscented Kalman filter [6], the cubature Kalman filter (CKF) [7], the quadrature Kalman filter [8] etc. demonstrate better stability and accuracy.…”

Section: Introductionmentioning

confidence: 99%

Gaussian sum pseudolinear Kalman filter for bearings‐only tracking

Jiang

Cai

2020

IET Control Theory & Applications

View full text Add to dashboard Cite

“…Nevertheless, PF has some inherent practical problems, such as complexity of calculation, selection strategy of importance function and so on. In the field of information theoretic learning, maximum correntropy (MC) criterion has been successfully utilized for the non-Gaussian signal processing problems [28][29][30][31][32][33][34][35][36][37][38][43][44][45][46]. Under the MC criterion, several effective filter design methods were also developed for the non-Gaussian systems.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear Non-Gaussian Estimation Using Maximum Correntropy Square Root Cubature Information Filtering

Feng

Zhou

et al. 2020

IEEE Access

View full text Add to dashboard Cite

This paper concerns the nonlinear filter designing methods in the information space of the nonlinear systems with non-Gaussian noises. Firstly, the prediction information vector is obtained by the traditional square root cubature information filtering algorithm. Then, under the maximum correntropy criterion, the prediction information vector is corrected with the contribution information vector obtained by the non-Gaussian measurement. The information filtering gain is obtained by utilizing the state information correntropy matrix and the measurement information correntropy matrix, in which, the state prediction is taken as the state value. In order to improve the advantage of the above nonlinear non-Gaussian information filter in filtering accuracy, with the help of fixed-point theory, an iterative computation method is further developed to update the estimation information vector and the state estimate. The effectiveness of the two proposed nonlinear non-Gaussian filtering methods is illustrated in final four simulation examples. INDEX TERMS Maximum correntropy criterion; information correntropy matrix; square root cubature information filter;

show abstract

Maximum correntropy sparse Gauss–Hermite quadrature filter and its application in tracking ballistic missile

Cited by 41 publications

References 25 publications

Robust filtering with randomly delayed measurements and its application to ballistic target tracking in boost phase

Robust filtering with randomly delayed measurements and its application to ballistic target tracking in boost phase

Gaussian sum pseudolinear Kalman filter for bearings‐only tracking

Nonlinear Non-Gaussian Estimation Using Maximum Correntropy Square Root Cubature Information Filtering

Contact Info

Product

Resources

About