本期导读

主题

交通控制、自适应控制、非线性系统、故障诊断、机器视觉、强化学习、多阶段优化、分布式控制、多智能体系统...

全球科研机构

美国New Jersey Institute of Technology；英国Newcastle University；中国科学院自动化研究所、浙江大学、上海交通大学、哈尔滨工业大学、南开大学、北京理工大学、中国科学技术大学、西安交通大学、大连理工大学、华东理工大学、重庆大学、四川大学；之江实验室...

X. H. Wen and M. C. Zhou, “Evolution and role of optimizers in training deep learning models,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2039–2042, Oct. 2024. doi: 10.1109/JAS.2024.124806 

Q. Ji, X. Wen, J. Jin, Y. Zhu, and  Y. Lv,  “Urban traffic control meets decision recommendation system: A survey and perspective,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2043–2058, Oct. 2024. doi: 10.1109/JAS.2024.124659 

> Presents a survey on decision recommendation systems in traffic management.

> Illustrates key components in traffic control decision recommendation systems.

> Highlights the use of human- and data-driven methods for traffic optimization.

Z. Li, Y. Wang, and  Y.  Song,  “Achieving given precision within prescribed time yet with guaranteed transient behavior via output based event-triggered control,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2059–2067, Oct. 2024. doi: 10.1109/JAS.2023.124134 

> For a class of uncertain nonlinear systems, the given control precision is achieved within a prescribed finite time.

> Burden of both the sensing and computation is largely reduced by designing the state filters and utilizing intermittent input signal.

> Initial condition restriction is removed by constructing a novel performance scaling function and an error transformation.

X. Li, S. Yu, Y. Lei, N. Li, and  B. Yang,  “Dynamic vision-based machinery fault diagnosis with cross-modality feature alignment,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2068–2081, Oct. 2024. doi: 10.1109/JAS.2024.124470 

> A contactless machine fault diagnosis method is proposed with dynamic vision.

> A cross-modality alignment method is proposed for vision and accelerometer data.

> An event erasing method is proposed to enhance model robustness.

W. Ren, Z.-R. Pan, W. Xia, and  X.-M. Sun,  “Hierarchical controller synthesis under linear temporal logic specifications using dynamic quantization,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2082–2098, Oct. 2024. doi: 10.1109/JAS.2024.124473 

> Hierarchical control structure to connect high-level plan with local-level control.

> Dynamic quantization based realization verification for LTL specifications.

> A novel local-to-global control strategy to reduce computational complexity greatly.

M. Wang, H. Yan, J. Qiu, and W. Ji, “Fuzzy-model-based finite frequency fault detection filtering design for two-dimensional nonlinear systems,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2099–2110, Oct. 2024. doi: 10.1109/JAS.2024.124452 

> Developed a fuzzy model based filtering synthesis method for Roesser type 2-D nonlinear systems with disturbances and faults.

> A frequency based fault detection filtering design method is proposed to generate a residual signal with both sensitivity to faults and robustness to external disturbances.

> An evaluation function together with its threshold has been designed, and then a finite frequency fault detection algorithm has been developed.

Z. Qiu, S. Wang, D. You, and  M. C. Zhou,  “Bridge bidding via deep reinforcement learning and belief Monte Carlo search,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2111–2122, Oct. 2024. doi: 10.1109/JAS.2024.124488 

> A cost-effective deep reinforcement learning approach for training a Bridge bidding agent is proposed.

> A novel search-based method which integrates a belief network to predict cards of other players and a policy network to evaluate candidate actions is proposed. 

> A tournament between trained Bridge bidding agents and WBridge5, an award-winning Bridge software is conducted. 

G. Li, B. Zhao, X. Su, D. Li, Y. Yang, Z. Zeng, and  L. Hu,  “Learning sequential and structural dependencies between nucleotides for RNA N6-methyladenosine site identification,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2123–2134, Oct. 2024. doi: 10.1109/JAS.2024.124233 

> Considering the non-Euclidean spatial properties in the 3D structures of RNA molecules, CR-NSSD explores the possibility of learning structural dependencies between nucleotides solely from their sequence information, and combines both structural and sequential dependencies for improved performance of RNA m6A modification site identification. The consideration of multi-view dependencies between nucleotides strengths the prediction ability of CR-NSSD by fully exploiting the RNA sequence information.

X. Zhang, Z. Han, and  J. Zhao,  “A multi-stage differential-multifactorial evolutionary algorithm for ingredient optimization in the copper industry,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2135–2153, Oct. 2024. doi: 10.1109/JAS.2023.124116 

> An ingredient optimization model considering the feeding stability is developed.

> MS-DME algorithm is proposed to optimize the model.

> Infeasible Ingredient lists are effectively repaired.

Y. Zhang, Z. Liu, and  Z. Chen,  “A PI+R control scheme based on multi-agent systems for economic dispatch in isolated BESSs,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2154–2165, Oct. 2024. doi: 10.1109/JAS.2024.124236 

> Reset mechanism is designed based on zero crossing. The parameter conditions for the reset mechanism to function are given.

> Considering capacity constraints, the convergence rate of the modified MC scheme by time trigger is also accelerated by the reset mechanism.

> Performance of the controller under Zeno behavior and input delay is analyzed.

S. Cong and  Z. Dong,  “Pure state feedback switching control based on the online estimated state for stochastic open quantum systems,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2166–2178, Oct. 2024. doi: 10.1109/JAS.2023.124071 

> First time to propose an OQST-SFC for pure state transfer of stochastic open quantum systems.

> One of on-line quantum state estimation algorithms and pure state transfer method of stochastic open quantum systems are combined to study the pure state switching feedback control based on online estimated state for stochastic open quantum systems.

> Focuses on analyzing the four different control cases when the initial estimated state and the controlled quantum system’s initial state are different in the proposed OQST-SFC strategy.

K. Xia, X. Li, K. Li, and Y. Zou, “Distributed predefined-time control for cooperative tracking of multiple quadrotor UAVs,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2179–2181, Oct. 2024. doi: 10.1109/JAS.2023.123861 

X. Wang, S. Zhao, M. Yang, X. Wang, and X. Wu, “Neural network-based state estimation for nonlinear systems with denial-of-service attack under try-once-discard protocol,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2182–2184, Oct. 2024. doi: 10.1109/JAS.2023.123690 

Y. Song, Y. Liu, and W. Zhao, “Approximately bi-similar symbolic model for discrete-time interconnected switched system,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2185–2187, Oct. 2024. doi: 10.1109/JAS.2023.123927 

H. Chen, M. Lin, J. Liu, and Z. Xu, “Scalable temporal dimension preserved tensor completion for missing traffic data imputation with orthogonal initialization,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 10, pp. 2188–2190, Oct. 2024. doi: 10.1109/JAS.2024.124278