本期导读

主题

控制性能监测、静态及动态分析、大型语言模型、软件安全、分布式优化、联合学习、鲁棒控制、卡尔曼滤波、多传感器系统、深度网络、迁移学习、DoS、自适应预测控制、非线性多智能体系统、自适应动态编程、智能电网、熵、时间序列预测、值迭代...

全球科研机构

英国Brunel University London；澳大利亚University of Adelaide、Swinburne University of Technology；加拿大University of Alberta、University of Victoria；中国科学院自动化研究所、浙江大学、同济大学、上海交通大学、东北大学、大连理工大学、武汉理工大学、南京邮电大学、武汉大学、东南大学、重庆大学、西南大学、西北工业大学、哈尔滨理工大学、广东工业大学；国科大杭州高等研究院、中国船舶科学研究中心、紫金山实验室...

W. Zheng and F.-Y. Wang, “Knowledge as not only justified true beliefs in vision,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 297–299, Feb. 2025. doi: 10.1109/JAS.2024.124584 

P. Song, J. Wang, C. Zhao, and  B. Huang,  “From static and dynamic perspectives: A survey on historical data benchmarks of control performance monitoring,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 300–316, Feb. 2025. doi: 10.1109/JAS.2024.124902 

> First survey on historical data benchmarks of control performance monitoring.

> Manifestations of control performance are revealed from static and dynamic aspects.

> A novel taxonomy is proposed for existing historical data benchmarks.

X. Zhu, W. Zhou, Q.-L. Han, W. Ma, S. Wen, and  Y. Xiang,  “When software security meets large language models: A survey,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 317–334, Feb. 2025. doi: 10.1109/JAS.2024.124971 

> Systematically demonstrates techniques using LLMs in software security.

> Breakdowns each technique and demonstrates where LLMs are used in each stage.

> Digs deep into the reasons and advantages of using LLMs in software security.

M. Wei, W. Yu, D. Chen, M. Kang, and  G. Cheng,  “Privacy distributed constrained optimization over time-varying unbalanced networks and its application in federated learning,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 335–346, Feb. 2025. doi: 10.1109/JAS.2024.124869 

> A privacy-preserving zeroth-order optimization algorithm for constrained distributed optimization problems over time-varying unbalanced graphs is proposed, named as DP-ZOCOA.

X. Chen, Y. Wang, and  Y. Song,  “Unifying fixed time and prescribed time control for strict-feedback nonlinear systems,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 347–355, Feb. 2025. doi: 10.1109/JAS.2024.124401 

> A prescribed-time filter is proposed to allow the upper bound of the convergence time for the estimation error to be provided precisely in advance.

> Proposed prescribed-time control algorithm enables the control gains not to escape to infinity and the system to operate on the whole time interval but not only on the finite time interval, distinguishing itself from most existing prescribed-time control methodologies based on time-varying gains.

J. Li, Z. Wang, J. Hu, H. Dong, and H. Liu, “Cubature Kalman fusion filtering under amplify-and-forward relays with randomly varying channel parameters,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 356–368, Feb. 2025. doi: 10.1109/JAS.2024.124590 

> A novel cubature Kalman fusion filtering algorithm is proposed.

> The amplify-and-forward relays are used to regulate signal communication.

> The randomly varying channel parameters are considered.

T. Wang, F. Zhou, Y. Wu, J. Zhao, and W. Wang, “A multi-condition sequential network ensemble for industrial energy storage prediction considering the condition switching characteristics,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 369–380, Feb. 2025. doi: 10.1109/JAS.2024.124962 

> In order to especially consider the time dependence of different working-conditions, a central-wise condition sequential network is developed.

> A branch combination transfer learning strategy is developed to tackle the sample imbalance problem of different condition data.

H. Xiong, G. Chen, H. Ren, and H. Li, “Broad-learning-system-based model-free adaptive predictive control for nonlinear MASs under DoS attacks,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 381–393, Feb. 2025. doi: 10.1109/JAS.2024.124929 

> A fully data-driven, model-free control method is proposed to address the containment control problem of NMASs under DoS attacks.

> To cope with DoS attacks, a prediction mechanism based on the broad learning system is designed within the model-free adaptive control framework.

> The trained broad learning system uses historical data to predict the lost output data due to DoS attacks.

X. Shi and  C. Sun,  “Penalty function-based distributed primal-dual algorithm for nonconvex optimization problem,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 394–402, Feb. 2025. doi: 10.1109/JAS.2024.124935 

> A distributed nonconvex optimization over undirected network is considered.

> A penalty-function method is provided to solve the nonconvex problem.

> An adaptive penalty factor is provided for the local constraints.

K. Xiong, Q. Wei, and  H. Li,  “Residential energy scheduling with solar energy based on Dyna adaptive dynamic programming,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 403–413, Feb. 2025. doi: 10.1109/JAS.2024.124809 

> An improved ADHDP method that incorporates learning and planning ideas.

> Dyna architecture to enhance learning efficiency and generalization performance.

> Improved utility function enhances renewable energy utilization.

X. Wang, D. Yu, and  X. Li,  “Impulsive consensus of MASs with input saturation and DoS attacks,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 414–424, Feb. 2025. doi: 10.1109/JAS.2024.124944 

> Considers the coupling of input saturation and denial of service (DoS) attacks.

> Input saturation with random impulsive signal intensity is analyzed.

> Linear matrix inequalities are used to characterize the coupling.

X. Chen, Z. Su, L. Jin, and  S. Li,  “A correntropy-based echo state network with application to time series prediction,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 425–435, Feb. 2025. doi: 10.1109/JAS.2024.124932 

> Proposes a correntropy-based echo state network.

> Presents an incremental learning algorithm for the proposed network.

> Conducts experiments on benchmark problems and comparisons with existing works to verify the effectiveness and superiority of the proposed network in predicting noisy and noise-free time series.

Y. Zhang, Y. Wang, and  Y. Cai,  “Value iteration-based distributed adaptive dynamic programming for multi-player differential game with incomplete information,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 436–447, Feb. 2025. doi: 10.1109/JAS.2024.124950 

> Presents a novel distributed VI-based ADP method for multi-player differential game models with incomplete information.

> Incomplete information structure is characterized by the limit information from neighbors.

> Each player completes the learning asynchronously and independently, although it does not know who among the others updates the policy and when it happens.

L. Xu, D. Xu, X. Yi, C. Deng, T. Chai, and T. Yang, “Decentralized federated learning algorithm under adversary eavesdropping,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 448–456, Feb. 2025. doi: 10.1109/JAS.2024.125079 

> Propose the TRADE concept for secure weight transmission.

> Prove similar convergence as the primal-dual SGD baseline.

> Validate privacy-preserving with an eavesdropper error bound.

Z.-X. Li, Y.-L. Wang, and F. Wang, “DI-YOLOv5: An improved dual-wavelet-based YOLOv5 for dense small object detection,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 457–459, Feb. 2025. doi: 10.1109/JAS.2024.124368 

Z. Feng and S. Yao, “Dynamic event-triggered active disturbance rejection formation control for constrained underactuated AUVs,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 460–462, Feb. 2025. doi: 10.1109/JAS.2024.124617 

J. Ding, M. Zheng, and H. Yu, “Soft resource slicing for industrial mixed traffic in 5G networks,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 463–465, Feb. 2025. doi: 10.1109/JAS.2024.124761 

T. Zhang, J. Cao, M. Abdel-Aty, and A. Kashkynbayev, “Finite-time stability of impulsive and switched hybrid systems with delay-dependent impulses,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 466–468, Feb. 2025. doi: 10.1109/JAS.2024.124758 

M. Yu and X. Li, “Exponential stability of impulsive system via saturated sliding mode control,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 469–471, Feb. 2025. doi: 10.1109/JAS.2024.124734

Z. Wang, H. Zhang, C. Yang, and X. Cao, “Improved zero-dynamics attack scheduling with state estimation,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 472–474, Feb. 2025. doi: 10.1109/JAS.2024.124737 

P. Tang and X. Luo, “Neural Tucker factorization,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 475–477, Feb. 2025. doi: 10.1109/JAS.2024.124977 

Y. Huang, G.-P. Liu, Y. Yu, and W. Hu, “Constrained networked predictive control for nonlinear systems using a high-order fully actuated system approach,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 2, pp. 478–480, Feb. 2025. doi: 10.1109/JAS.2024.124764