Article
Automation & Control Systems
Liang-Qun Li, Ying-chun Sun, Zong-Xiang Liu
Summary: A novel maximum fuzzy correntropy Kalman filter (MFC-KF) algorithm is proposed to optimize the Kalman filter by reducing the effect of common correntropy and applying a new optimization criterion. Combined with the least-squares method and adaptive kernel width setting, the proposed algorithm can track a target more accurately than other traditional Kalman filters.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2021)
Article
Chemistry, Analytical
Mahendra Mallick, Xiaoqing Tian, Yun Zhu, Mark Morelande
Summary: This study considers the state estimation of a maneuvering target in 3D using bearing and elevation measurements from a passive infrared search and track sensor. The target moves with nearly constant turn in the XY-plane and nearly constant velocity along the Z-axis. The first and second-order Taylor approximations are used to discretize the continuous-time model, and the cubature Kalman filter is employed for state estimation. Numerical results show that the second-order Taylor approximation provides the best accuracy using either polar velocity or Cartesian velocity-based models.
Article
Automation & Control Systems
Haonan Jiang, Yuanli Cai, Zhenhua Yu
Summary: This work explores the observability problem in bearings-only tracking (BOT) and proposes a series of metrics based on the condition number to quantitatively describe the observability degree and evaluate the tracking performance. The metrics can also be applied to sensor trajectory optimization and sensor configuration to enhance the tracking performance.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Geochemistry & Geophysics
Lu Shen, Hongtao Su, Ze Li, Congyue Jia, Ruixing Yang
Summary: This article introduces two nonlinear trackers based on the Transformer for smoothing, filtering, and predicting target states in nonlinear target tracking. The proposed algorithms improve the efficiency and accuracy of inference and reduce computational load compared to traditional methods and recurrent neural network-based methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Automation & Control Systems
Zhijin Chen, Branko Ristic, Jeremie Houssineau, Du Yong Kim
Summary: This paper focuses on bearings-only tracking using passive sensors, adopting possibility functions to represent uncertainties instead of probability distributions. It explores the design of reward functions based on possibility theory and shows that the proposed framework outperforms the Bayesian probabilistic framework in the presence of model mismatch.
Article
Environmental Sciences
Zihao Huang, Shijin Chen, Chengpeng Hao, Danilo Orlando
Summary: In bearings-only target tracking, the pseudo-linear Kalman filter (PLKF) is popular for stability and low computational burden, but suffers from bias problems due to correlated measurement vector and noise; an unbiased PLKF algorithm (UB-PLKF) is proposed to address this issue, along with a velocity-constrained version (VC-PLKF) to further improve performance, outperforming other methods in both non-manoeuvring and manoeuvring scenarios according to simulations.
Article
Engineering, Electrical & Electronic
Mohsen Ebrahimi, Mahdi Ardeshiri, Sedigheh Alaie Khanghah
Summary: This paper reviews and analyzes the maneuvering target tracking model and proposes an improved algorithm for accurately estimating the target's state in the presence of measurement noise. The proposed method uses the multiple-model Interacting Multiple Model algorithm and introduces higher-order Markov models to describe the system behavior. The results show that the algorithm performs well in target tracking.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Engineering, Electrical & Electronic
Shan Zhong, Bei Peng, Lingqiang Ouyang, Xinyue Yang, Hongyu Zhang, Gang Wang
Summary: This article presents a framework for a pseudolinear Kalman filter (PLKF) based on the maximum correntropy criterion for the bearings-only target tracking problem in non-Gaussian environments. The proposed estimation method, including several algorithms, outperforms the traditional Kalman filter in non-Gaussian noise environments.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
S. Koteswara Rao, M. Kavitha Lakshmi, Kausar Jahan, G. Naga Divya, B. Omkar Lakshmi Jagan
Summary: This research focuses on the evolution of acceptance criterion for target motion parameter estimation in actual scenarios. Mathematical modeling is used along with an unscented Kalman filter to estimate the parameters. The acceptance criterion is derived based on the covariance matrix and standard deviation of errors in the estimated parameters.
IETE JOURNAL OF RESEARCH
(2023)
Article
Engineering, Electrical & Electronic
Wen Zhang, Xuanzhi Zhao, Zengli Liu, Kang Liu, Bo Chen
Summary: Handling the nonlinearity between the measurement and kinematic states is the core issue in target tracking based on radar or sonar. This paper proposes a new filter with a linear structure to achieve nonlinear tracking by integrating information in the polar coordinate system. The new filter effectively improves the tracking accuracy, and different posterior Cramer- Rao lower bounds (PCRLBs) for fusion estimation in Cartesian coordinates and polar coordinates are given and compared.
Article
Mathematics, Interdisciplinary Applications
Xiong Kai, Wei Chunling
Summary: This paper proposes a Q-learning-based target selection algorithm for spacecraft autonomous navigation to enhance the performance of the extended Kalman filter. Numerical simulations demonstrate that this algorithm outperforms traditional target selection strategies.
JOURNAL OF SYSTEMS SCIENCE & COMPLEXITY
(2021)
Article
Computer Science, Information Systems
Yuankai Li, Jiaxin Lou, Xiaosu Tan, Yanke Xu, Jinpeng Zhang, Zhongliang Jing
Summary: Kernel method is a non-parametric linearization method that utilizes nonlinear projection and kernel function to improve state estimation in uncertain systems. An adaptive kernel learning Kalman filtering method is proposed and applied to maneuvering target tracking, demonstrating its effectiveness in model-free tracking.
Article
Engineering, Aerospace
Yingjie Zhang, Jian Lan, Mahendra Mallick, X. Rong Li
Summary: The article introduces a new filtering method based on uncorrelated conversion, which can effectively perform nonlinear filtering. The constructed pseudomeasurement is uncorrelated with the original measurement, improving the efficiency of information utilization. Simulation results demonstrate that this method has better estimation accuracy compared to traditional particle filters at nearly the same computational cost.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2021)
Article
Chemistry, Analytical
Guanghui Zhao, Zelin Wang, Yixiong Huang, Huirong Zhang, Xiaojing Ma
Summary: In this study, a transformer-based network (TBN) is proposed to track maneuvering targets. With the attention mechanism and center-max normalization, TBN can capture the long short-term dependencies of target states globally while reducing training complexity and improving generalization.
Article
Chemistry, Multidisciplinary
Yalun Luo, Zhaoming Li, Yurong Liao, Haining Wang, Shuyan Ni
Summary: This paper proposes a strong tracking cubature Kalman filter adaptive interactive multi-model algorithm for the tracking of hypersonic targets. By introducing fading factors and singular value decomposition, the algorithm improves tracking accuracy and convergence speed.
APPLIED SCIENCES-BASEL
(2022)