Article
Automation & Control Systems
Yulong Huang, Yonggang Zhang, Peng Shi, Jonathon Chambers
Summary: This article introduces a new variational adaptive Kalman filter using a Gaussian-inverse-Wishart mixture distribution for linear systems with partially unknown state and measurement noise covariance matrices. By establishing a hierarchical Gaussian model, the system state vector and noise covariance matrices are jointly estimated. Examples are provided to demonstrate the effectiveness and potential of this new filtering design technique.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Congcong Ye, Jixiang Yang, Han Ding
Summary: This paper introduces a closed-loop framework of LfD based on the bagging method of Gaussian Mixture Model and Gaussian Mixture Regression, which divides original demonstration data into multiple sub-training data for robust and high precision reproduction. Experimental results show that the proposed method can significantly meet task constraints without increasing algorithm complexity.
JOURNAL OF INTELLIGENT MANUFACTURING
(2022)
Article
Computer Science, Interdisciplinary Applications
Yinlin Fu, Xiaonan Liu, Suryadipto Sarkar, Teresa Wu
Summary: A new algorithm called Expectation Selection Maximization (ESM) is proposed in this paper to address the issue of confusion and increased computational cost in GMM models by adding a feature selection step. The introduction of a relevancy index (RI) assists in feature selection by indicating the probability of assigning data points to specific clustering groups. The theoretical analysis justifies the effectiveness of RI for feature selection.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Information Systems
Kevin R. Ford, Anton J. Haug
Summary: This study provides a concise derivation of the probability density function (PDF) for bearing in tracking a two-dimensional Cartesian state of a target using polar observations, and explores the limiting behavior of this distribution while parameterizing the target range.
Review
Engineering, Mechanical
Xiaoxu Wang, Haoran Cui, Tiancheng Li, Yan Liang, Zhengtao Ding
Summary: A new variational Gaussian regression filter is proposed in this paper by incorporating variational parameters into a linear parametric Gaussian regression process. The filtering evidence lower bound serves as a quantitative evaluation rule for different filters, and a relationship between F-ELBO and M-ELBO is identified. The accuracy performance improvement of VGRF can be theoretically explained based on these findings.
NONLINEAR DYNAMICS
(2021)
Article
Automation & Control Systems
Haoran Cui, Long Zhang, Xiaoxu Wang, Mingyong Liu, Binglu Wang
Summary: In this paper, a novel iterative nonlinear filter based on the variational Bayesian framework is proposed, which can achieve highly accurate state estimation and numerical stability in nonlinear systems.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2023)
Article
Computer Science, Interdisciplinary Applications
Mahshad Valipour, Luis A. Ricardez-Sandoval
Summary: Studies on Extended Kalman Filter that explicitly consider non-Gaussian process uncertainties are scarce. In the current study, a novel approach called Abridged Gaussian Sum Extended Kalman Filter (AGS-EKF) is developed to improve EKF performance for such systems. AGS-EKF offers higher accuracy of estimation in shorter computational times by considering overall Gaussian mixture of process uncertainties.
COMPUTERS & CHEMICAL ENGINEERING
(2021)
Article
Environmental Sciences
Zhen Gao, Kun Fang, Zhipeng Wang, Kai Guo, Yuan Liu
Summary: This paper introduces an ionosphere-free (Ifree) filtering algorithm for ensuring the integrity of a ground-based augmentation system (GBAS). It proposes an overbounding framework based on a Gaussian mixture model (GMM) to handle the errors outputted by the Ifree algorithm. The performance of the algorithm is evaluated through Monte Carlo simulations and real-world road tests.
Article
Energy & Fuels
Peng Guo, Wentao Ma, Dele Yi, Xinghua Liu, Xiaofei Wang, Lujuan Dang
Summary: This paper proposes a novel robust state estimation method that enhances the robustness of the square-root cubature Kalman filter by incorporating a mixture correntropy loss. The method overcomes the issue of non-Gaussian measurement noise interference and achieves high estimation accuracy in different cases.
JOURNAL OF ENERGY STORAGE
(2023)
Article
Geochemistry & Geophysics
Afonso M. Teodoro, Jose M. Bioucas-Dias, Mario A. T. Figueiredo
Summary: This article introduces a denoising method for hyperspectral images that takes into account both spatial and spectral features, achieving high-quality image reconstruction under Gaussian and Poissonian noise, outperforming other state-of-the-art methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Mathematics
Angel L. Cedeno, Rodrigo A. Gonzalez, Boris I. Godoy, Rodrigo Carvajal, Juan C. Aguero
Summary: This paper provides a comprehensive overview of the state-of-the-art algorithms for state estimation subject to quantized measurements, and performs an exhaustive comparison among them. The evaluation criteria include the accuracy of state estimation, dimensionality issues, hyperparameter selection, user friendliness, and computational cost. Extensive numerical simulations, including a practical application, are conducted to analyze the accuracy and computational cost.
Article
Engineering, Marine
Sisi Wang, Lijun Wang, Namkyun Im, Weidong Zhang, Xijin Li
Summary: A real-time parameter identification method based on nonlinear Gaussian filtering algorithm and nonlinear ship response model is proposed to improve system identification accuracy and reduce computational complexity.
Article
Engineering, Marine
Doudou Li, Ron Patton
Summary: This paper presents a model predictive velocity tracking control method based on a hierarchical structure for a Wavestar-like device in the WEC-SIM benchmark, aiming to achieve energy-maximizing control and reduce the levelized cost of energy (LCOE). The proposed method estimates the wave excitation moment (WEM) using a Kalman filter and obtains the amplitude and angular frequency of the WEM using an extended Kalman filter (EKF) to compute the reference velocity. A low-level model predictive control (MPC) method is then designed to ensure the wave energy converter (WEC) tracks the optimal reference velocity for maximum energy extraction. Two Gaussian Process (GP) models are used to predict the future WEM and reference velocity for the MPC design. This study provides a new perspective for energy-maximizing tracking control based on model predictive control.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Multidisciplinary Sciences
Claudio Urrea, Rayko Agramonte
Summary: This study presents the applications of the Kalman filter in robotics and highlights its importance. It also discusses the improvements made to enhance its performance. Experimental results show that the least squares and unscented Kalman filter methods yield good parameter identification results, and the robot's performance can be compared using different performance indexes.
Article
Computer Science, Information Systems
Dah-Jing Jwo, Chien-Hao Tseng
Summary: This paper evaluates the state estimation performance of processing nonlinear/non-Gaussian systems using the cubature particle filter (CPF), which combines the cubature Kalman filter (CKF) and the particle filter (PF). The CPF algorithm effectively resolves nonlinear/non-Gaussian problems by using the CKF to generate the importance density function. The CPF can achieve a maximum a posteriori probability estimate of the nonlinear system, improving estimation accuracy compared to other particle filter and Kalman filter based approaches.
CMC-COMPUTERS MATERIALS & CONTINUA
(2021)