Jiayuan Yan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jiayuan Yan has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Zhi-Hong Guan
Huazhong University of Science and Technology
9 shared publications - Bin Hu
South China University of Technology
8 shared publications - Zhengquan Chen
School of Electrical Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China
5 shared publications - Yong Jin
School of Artificial Intelligence, Henan University, Zhengzhou, China
4 shared publications Ding-Xue Zhang
3 shared publications- Lei Shi
Henan University
3 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (21) , 51358-51372
On Controllability of IoT-Enabled Impulse-Effected Networked Multiagent Systems Under Continuous and Discrete Communications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (21) , 51358-51372
On Controllability of IoT-Enabled Impulse-Effected Networked Multiagent Systems Under Continuous and Discrete Communications
- Open Access
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 10013-10043
A Multisemantic Multipartition Dual-Layer Alternating Joint Update Markov Random Field Method for Remote Sensing Image Semantic Segmentation
Semantic segmentation of remote sensing images is an important research direction in remote sensing science. With the increasing resolution of remote sensing imagery, precise semantic segmentation faces the critical challenge of balancing the segmentation of small targets within large-scale remote sensing images. While Markov Random Field (MRF) models have shown promise in spatial modeling, existing multi-layer approaches often suffer from unidirectional information transfer and insufficient feature interaction. Consequently, they struggle to preserve small-scale objects against strong global statistical constraints, leading to smoothed-out boundaries and incomplete information transmission. To address these issues, this paper proposes a Multi-Semantic Multi-Partition Dual-Layer Alternating Joint Update Markov Random Field method (MSMP-DLMRF). We construct a heterogeneous dual-layer framework. The first layer is a coarse-grained high-semantic original layer for global consistency, while the second layer is a fine-grained low-semantic partition layer designed to increase the proportion of small-object areas. Crucially, an alternating joint iterative update mechanism is designed, and a dynamic bidirectional feedback loop is established, so that the exploitation of spatial–feature–semantic relationships across layers is maximized. The proposed method was tested on texture images, GeoEye, LoveDA, Gaofen Image Dataset (GID) and ISPRS Potsdam datasets. Extensive experimental results demonstrate that the proposed method achieves superior segmentation performance compared to selected methods. Specifically, MSMP-DLMRF exhibits significant advantages in recognizing small-scale targets and maintaining boundary integrity in complex scenes.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19 (19391404) , 10013-10043
Open AccessA Multisemantic Multipartition Dual-Layer Alternating Joint Update Markov Random Field Method for Remote Sensing Image Semantic Segmentation
Semantic segmentation of remote sensing images is an important research direction in remote sensing science. With the increasing resolution of remote sensing imagery, precise semantic segmentation faces the critical challenge of balancing the segmentation of small targets within large-scale remote sensing images. While Markov Random Field (MRF) models have shown promise in spatial modeling, existing multi-layer approaches often suffer from unidirectional information transfer and insufficient feature interaction. Consequently, they struggle to preserve small-scale objects against strong global statistical constraints, leading to smoothed-out boundaries and incomplete information transmission. To address these issues, this paper proposes a Multi-Semantic Multi-Partition Dual-Layer Alternating Joint Update Markov Random Field method (MSMP-DLMRF). We construct a heterogeneous dual-layer framework. The first layer is a coarse-grained high-semantic original layer for global consistency, while the second layer is a fine-grained low-semantic partition layer designed to increase the proportion of small-object areas. Crucially, an alternating joint iterative update mechanism is designed, and a dynamic bidirectional feedback loop is established, so that the exploitation of spatial–feature–semantic relationships across layers is maximized. The proposed method was tested on texture images, GeoEye, LoveDA, Gaofen Image Dataset (GID) and ISPRS Potsdam datasets. Extensive experimental results demonstrate that the proposed method achieves superior segmentation performance compared to selected methods. Specifically, MSMP-DLMRF exhibits significant advantages in recognizing small-scale targets and maintaining boundary integrity in complex scenes.
- 17 November 2025
- Published by Elsevier in Journal of the Franklin Institute
- Vol. 362 (18)
A Robust Iterative learning technique-based fault detection for nonlinear singular systems with disturbances and parameter uncertainties
- 17 November 2025
- Published by Elsevier in Journal of the Franklin Institute
- Vol. 362 (18)
A Robust Iterative learning technique-based fault detection for nonlinear singular systems with disturbances and parameter uncertainties
- 17 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 71 (4) , 10580-10590
Observability Analysis on Discrete-Time Matrix-Weighted Multi-Agent Systems
Unmanned systems are advancing rapidly, driving widespread attention to multi-agent systems composed of clustered consumer electronic devices. Accurately and timely estimating component states of multi-agent systems during signal transmission is critical for performance. However, existing studies primarily focus on observability in scalar-weighted networks, neglecting matrix-weighted alternatives. These matrix-weighted networks provide greater structural complexity and enhanced modeling capabilities for real-world discrete-time communication models. This paper addresses the observability gap in discrete-time multi-agent systems over matrix-weighted networks, a domain neglected in existing scalar-weighted studies. The observability conditions for systems with fixed weights are first established. The observability criteria are established via spectral decomposition and graph theory, analyzing two eigenvalue distribution cases of the system state matrix. Subsequently, observability criteria for systems with time-varying weights are derived. The criteria are derived using algebraic graph theory, yielding graphical and algebraic conditions based on specific topological configurations. Computational studies validate the effectiveness of the proposed conditions.
- 17 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 71 (4) , 10580-10590
Observability Analysis on Discrete-Time Matrix-Weighted Multi-Agent Systems
Unmanned systems are advancing rapidly, driving widespread attention to multi-agent systems composed of clustered consumer electronic devices. Accurately and timely estimating component states of multi-agent systems during signal transmission is critical for performance. However, existing studies primarily focus on observability in scalar-weighted networks, neglecting matrix-weighted alternatives. These matrix-weighted networks provide greater structural complexity and enhanced modeling capabilities for real-world discrete-time communication models. This paper addresses the observability gap in discrete-time multi-agent systems over matrix-weighted networks, a domain neglected in existing scalar-weighted studies. The observability conditions for systems with fixed weights are first established. The observability criteria are established via spectral decomposition and graph theory, analyzing two eigenvalue distribution cases of the system state matrix. Subsequently, observability criteria for systems with time-varying weights are derived. The criteria are derived using algebraic graph theory, yielding graphical and algebraic conditions based on specific topological configurations. Computational studies validate the effectiveness of the proposed conditions.
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 37407-37418
Multisensor Management Method Based on Multistep Prediction of Bidirectional Joint Risk
In a multi-sensor collaborative tracking system, rational multi-sensor management methods can achieve optimal system performance. However, the complexity and variability of environmental risks, will lead to the accuracy and safety of tracking system reduce. Therefore, this paper proposes a multi-sensor management method based on multi-step prediction of bidirectional joint risk to rationally allocate limited sensor resources. Firstly, the paper comprehensively considers three risks including the radiation risk of our multi-sensors, the risk of detection loss, and the threat risk of opposing targets, meanwhile constructs a bidirectional joint risk model. Secondly, adaptive weights for the three risks are proposed to adjust the three risks in the above model. Then, based on the framework of time series prediction, the bidirectional joint risk are predicted. Finally, based on this, the problem of minimizing multi-step prediction bidirectional joint risk is proposed, and then achieving the rational allocation of multi-sensor resources. The simulation results demonstrate that the propose method is feasible, as it can effectively allocate limited sensor resources in a multi-risk environment, improving the accuracy and security of the tracking system.
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 25 (19) , 37407-37418
Multisensor Management Method Based on Multistep Prediction of Bidirectional Joint Risk
In a multi-sensor collaborative tracking system, rational multi-sensor management methods can achieve optimal system performance. However, the complexity and variability of environmental risks, will lead to the accuracy and safety of tracking system reduce. Therefore, this paper proposes a multi-sensor management method based on multi-step prediction of bidirectional joint risk to rationally allocate limited sensor resources. Firstly, the paper comprehensively considers three risks including the radiation risk of our multi-sensors, the risk of detection loss, and the threat risk of opposing targets, meanwhile constructs a bidirectional joint risk model. Secondly, adaptive weights for the three risks are proposed to adjust the three risks in the above model. Then, based on the framework of time series prediction, the bidirectional joint risk are predicted. Finally, based on this, the problem of minimizing multi-step prediction bidirectional joint risk is proposed, and then achieving the rational allocation of multi-sensor resources. The simulation results demonstrate that the propose method is feasible, as it can effectively allocate limited sensor resources in a multi-risk environment, improving the accuracy and security of the tracking system.
- 28 July 2025
- Published by Elsevier in Neural Networks
- Vol. 192, 107910
State bounding for discrete-time switched genetic regulatory networks with time delay and exogenous disturbances
- 28 July 2025
- Published by Elsevier in Neural Networks
- Vol. 192, 107910
State bounding for discrete-time switched genetic regulatory networks with time delay and exogenous disturbances
- 15 July 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 326
On controllability of discrete-time multi-agent networks based on impulsive and switching systems
- 15 July 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 326
On controllability of discrete-time multi-agent networks based on impulsive and switching systems
- 12 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (9) , 7287-7298
Neural Network–Based Reinforcement Iterative Learning Fault Estimation Scheme for Nonlinear Uncertain Manipulator Systems With Time-Delay
This article investigates the problem of fault estimation in nonlinear uncertain manipulator systems with time-delay. A novel fault estimation scheme is proposed, which optimizes the iterative learning (IL) estimator performance using a neural network (NN)-based reinforcement learning (RL) approach. Specifically, first, with the purpose of enhancing robustness and adaptability of RL, a new adaptive exponential reward function is designed. Then, to improve the performance of fault estimation, speed and accuracy are designed as optimization objectives. Simultaneously, by leveraging the IL estimator, the NN is continuously optimized in each iteration process to mitigate the issues of gradient vanishing and explosion. Further, by incorporating the H$\mathrm{\infty }$ performance index into the observer, an asymptotically convergent estimated error can be attained. Finally, numerical simulations are conducted to demonstrate the effectiveness of our method.
- 12 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21 (9) , 7287-7298
Neural Network–Based Reinforcement Iterative Learning Fault Estimation Scheme for Nonlinear Uncertain Manipulator Systems With Time-Delay
This article investigates the problem of fault estimation in nonlinear uncertain manipulator systems with time-delay. A novel fault estimation scheme is proposed, which optimizes the iterative learning (IL) estimator performance using a neural network (NN)-based reinforcement learning (RL) approach. Specifically, first, with the purpose of enhancing robustness and adaptability of RL, a new adaptive exponential reward function is designed. Then, to improve the performance of fault estimation, speed and accuracy are designed as optimization objectives. Simultaneously, by leveraging the IL estimator, the NN is continuously optimized in each iteration process to mitigate the issues of gradient vanishing and explosion. Further, by incorporating the H$\mathrm{\infty }$ performance index into the observer, an asymptotically convergent estimated error can be attained. Finally, numerical simulations are conducted to demonstrate the effectiveness of our method.
- 27 May 2025
- Published by Springer Nature in Nonlinear Dynamics
- p. 1-19
Robust fault detection and isolation for fuzzy Markov jump systems based on geometric approach
- 27 May 2025
- Published by Springer Nature in Nonlinear Dynamics
- p. 1-19
Robust fault detection and isolation for fuzzy Markov jump systems based on geometric approach
- Top Cited Papers
- 7 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (8) , 5561-5567
Reaching Opinion Consensus on Leader–Follower Signed Social Networks With Communication Delays
Consensus is a common opinion evolution phenomenon in social networks. Existing studies generally believe that opinion consensus is easy to reach when the network is connected and only trust relationships exist among agents. Distinguishing from existing studies, this paper aims to explore how to reach opinion consensus on leader-follower signed social networks with both trust and distrust relationships among agents. With the help of theoretical tools including binary relation composition and products of nonnegative matrices, the inequality constraints for trust and distrust degrees are established, which ensures that trust relationships can counteract the negative effects of distrust relationships on opinion consensus. In addition, further analysis is conducted on the factors that affect the convergence rate of opinion consensus, including the longest positive distance from leader to the followers as well as the upper bound of communication delays. At last, numerical examples are provided to validate the results.
- 7 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 70 (8) , 5561-5567
Top Cited PapersReaching Opinion Consensus on Leader–Follower Signed Social Networks With Communication Delays
Consensus is a common opinion evolution phenomenon in social networks. Existing studies generally believe that opinion consensus is easy to reach when the network is connected and only trust relationships exist among agents. Distinguishing from existing studies, this paper aims to explore how to reach opinion consensus on leader-follower signed social networks with both trust and distrust relationships among agents. With the help of theoretical tools including binary relation composition and products of nonnegative matrices, the inequality constraints for trust and distrust degrees are established, which ensures that trust relationships can counteract the negative effects of distrust relationships on opinion consensus. In addition, further analysis is conducted on the factors that affect the convergence rate of opinion consensus, including the longest positive distance from leader to the followers as well as the upper bound of communication delays. At last, numerical examples are provided to validate the results.