欺骗攻击和切换拓扑下多具空间动态特性神经网络的事件触发跟踪控制

操延燚; 孙亚平; 操钰婷; 朱明; 施开波; 黄廷文

doi:10.16383/j.aas.c260084

欺骗攻击和切换拓扑下多具空间动态特性神经网络的事件触发跟踪控制

doi: 10.16383/j.aas.c260084 cstr: 32138.14.j.aas.c260084

操延燚^{1, 2,},
孙亚平^3,,
操钰婷^4,,
朱明^{1, 2,},
施开波^{1, 2,},
黄廷文^4,

1.
成都大学电子信息与电气工程学院成都 610106
2.
四川省智能康复感知与控制国际联合研究中心成都 610106
3.
四川大学电子信息学院成都 610065
4.
深圳理工大学计算机科学与人工智能学院深圳 518107

基金项目: 国家自然科学基金(6250648, 62303336, 62506239, 62576223, 62366034), 四川省科技计划(2024NSFSC2056), 特殊环境机器人技术四川省重点实验室开放基金(24kftk01)资助

详细信息

作者简介:
操延燚：成都大学电子信息与电气工程学院副研究员. 2024年获中山大学计算机科学与技术博士学位. 主要研究方向为神经网络、多智能体系统和智能控制. E-mail: caoyanyi@cdu.edu.cn

孙亚平：四川大学电子信息学院副研究员. 2021年获华中科技大学博士学位. 主要研究方向为多智能体系统协调控制及其在无人系统集群上的应用. E-mail: sunyaping1010@163.com

操钰婷：深圳理工大学计算机科学与人工智能学院博士后. 2022年获湖南大学博士学位. 主要研究方向为非线性系统动力性和机器学习. 本文通信作者. E-mail: caoyuting09@126.com

朱明：成都大学电子信息与电气工程学院教授. 主要研究方向为信号处理, 模式识别和进化计算. E-mail: zhuming@cdu.edu.cn

施开波：成都大学电子信息与电气工程学院教授. 主要研究方向为神经网络、智能电网和多智能体系统. E-mail: skbs111@163.com

黄廷文：深圳理工大学计算机科学与人工智能学院讲席教授. 主要研究方向为非线性系统动力性、智能控制和多智能体系统. E-mail: huangtingwen@suat-sz.edu.cn

计量
- 文章访问数: 14
- HTML全文浏览量: 11
- 被引次数: 0
出版历程
- 收稿日期: 2026-01-29
- 录用日期: 2026-03-26
- 网络出版日期: 2026-08-28

Event-triggered Tracking Control for Multiple Neural Networks With Spatial Dynamic Characteristic Under Deception Attacks and Switching Topologies

CAO Yan-Yi^{1, 2
,},
SUN Ya-Ping^3
,,
CAO Yu-Ting^4
,,
ZHU Ming^{1, 2
,},
SHI Kai-Bo^{1, 2
,},
HUANG Ting-Wen^4
,

1.
School of Electronic Information and Electrical Engineering, Chengdu University, Chengdu 610106
2.
International Joint Research Center for Perception and Control of Intelligent Rehabilitation Systems of Sichuan Province, Chengdu 610106
3.
College of Electronics and Information Engineering, Sichuan University, Chengdu 610065
4.
Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology, Shenzhen 518107

Funds: Supported by National Natural Science Foundation of China (6250648, 62303336, 62506239, 62576223, 62366034), Sichuan Science and Technology Program (2024NSFSC2056), and Robot Technology Used for Special Environment Key Laboratory of Sichuan Province (24kftk01)

More Information

Author Bio:
CAO Yan-Yi　Associate professor at the School of Electronic Information and Electrical Engineering, Chengdu University. He received his Ph.D. degree in computer science and technology from Sun Yat-sen University in 2024. His research interests include neural networks, multi-agent systems, and intelligent control

SUN Ya-Ping　Associate professor at the College of Electronics and Information Engineering, Sichuan University. She received her Ph.D. degree from Huazhong University of Science and Technology in 2021. Her research interests include cooperative control of multi-agent systems and its application in unmanned system swarms

CAO Yu-Ting　Postdoctor at the Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology. She received her Ph.D. degree from Hunan University in 2022. Her research interests include dynamics of nonlinear systems and machine learning. Corresponding author of this paper

ZHU Ming　Professor at the School of Electronic Information and Electrical Engineering, Chengdu University. His research interests include signal processing, pattern recognition, and evolutionary computation

SHI Kai-Bo　Professor at the School of Electronic Information and Electrical Engineering, Chengdu University. His research interests include neural networks, smart grids, and multi-agent systems

HUANG Ting-Wen　Chair professor at the Faculty of Computer Science and Artificial Intelligence, Shenzhen University of Advanced Technology. His research interests include dynamics of nonlinear systems, intelligent control, and multi-agent systems

摘要

摘要: 神经网络在具身智能领域发挥了重要作用, 而其动态特性控制是其应用的关键. 首先, 针对欺骗攻击和切换拓扑给多具空间动态特性神经网络控制带来的挑战, 提出一种新的分布式事件触发跟踪同步控制协议. 然后, 通过构造新的Lyapunov函数和结合不等式技术, 得到多反应扩散神经网络系统在欺骗攻击和切换拓扑下的事件触发跟踪同步控制条件. 其次, 通过证明任意相邻事件的触发间隔存在正的下界, 排除了系统中的芝诺现象. 同时, 给出系统在无攻击时的事件触发跟踪同步控制条件. 此外, 控制器的参数由本文所得出的结论来确定, 进一步加强了系统的安全性. 最后, 给出一个数值仿真验证本文所设计控制策略的有效性.
- 反应扩散神经网络 /
- 欺骗攻击 /
- 切换拓扑 /
- 跟踪同步 /
- 事件触发控制
Abstract: Neural networks have played a significant role in the field of embodied intelligence, and the control of their dynamic characteristics is the key to their application. Firstly, a new distributed event-triggered tracking synchronization control protocol is proposed to address the challenges brought by deception attacks and switching topologies to the control of multiple neural network with spatial dynamic characteristic. Then, by constructing a new Lyapunov function and combining inequality techniques, the event-triggered tracking synchronization control conditions for multiple reaction-diffusion neural networks (MRDNNs) systems under deception attacks and switching topologies are obtained. Secondly, by proving that there exists a positive lower bound for the trigger interval of any adjacent events, the Zeno phenomenon in the system is excluded. Meanwhile, the event-triggered tracking synchronization control condition for the system without attacks is given. In addition, the parameters of the controller are determined by the criteria obtained in this paper, further enhancing the security of the system. Finally, a numerical simulation is provided to verify the effectiveness of the control strategy designed in this paper.
- Reaction-diffusion neural networks /
- deception attacks /
- switching topologies /
- tracking synchronization /
- event-triggered control

HTML全文

图 1 MRDNNs的切换拓扑结构

Fig. 1 The switching topologies of MRDNNs

下载: 全尺寸图片幻灯片

图 2 领导者神经网络的状态轨迹

Fig. 2 State trajectories of the leader neural network

下载: 全尺寸图片幻灯片

图 3 跟踪同步误差$ \|e_i(\cdot,\; t)\| $的演化过程(i=1, 2, 3, 4)

Fig. 3 Evolution process of the tracking synchronization error $ \|e_i(\cdot,\; t)\| $ (i=1, 2, 3, 4)

下载: 全尺寸图片幻灯片

图 4 采样同步误差$ \|e_i(\cdot,\; t_{k})\| $的演化过程(i=1, 2, 3, 4)

Fig. 4 Evolution process of the sampled synchronization error $ \|e_i(\cdot,\; t_{k})\| $ (i=1, 2, 3, 4)

下载: 全尺寸图片幻灯片

图 5 触发时间和触发间隔

Fig. 5 Triggering time and triggering interval

下载: 全尺寸图片幻灯片

参考文献(28)

[1]	孙长银, 袁心, 王远大, 柳文章. 具身智能自主无人系统技术. 自动化学报, 2025, 51(4): 762−777 Sun Chang-Yin, Yuan Xin, Wang Yuan-Da, Liu Wen-Zhang. Embodied intelligence autonomous unmanned systems technology. Acta Automatica Sinica, 2025, 51(4): 762−777
[2]	Wen G H, Zheng W X, Wan Y. Distributed robust optimization for networked agent systems with unknown nonlinearities. IEEE Transactions on Automatic Control, 2023, 68(9): 5230−5244 doi: 10.1109/TAC.2022.3216965
[3]	Driess D, Xia F, Sajjadi M S M, Lynch C, Chowdhery A, Ichter B, et al. PaLM-E: An embodied multimodal languagemodel. In: Proceedings of the 40th International Conference on Machine Learning. Honolulu, USA: ICML, 2023. 8469−8488
[4]	王文晟, 谭宁, 黄凯, 张雨浓, 郑伟诗, 孙富春. 基于大模型的具身智能系统综述. 自动化学报, 2025, 51(1): 1−19 doi: 10.16383/j.aas.c240542 Wang Wen-Sheng, Tan Ning, Huang Kai, Zhang Yu-Nong, Zheng Wei-Shi, Sun Fu-Chun. Embodied intelligence systems based on large models: A survey. Acta Automatica Sinica, 2025, 51(1): 1−19 doi: 10.16383/j.aas.c240542
[5]	Wen G H, Fu J J, Wang B W, Wang X D, Shen H. Robust formation tracking of ASVs with nonconservative collision avoidance: A control barrier function-based approach. IEEE Transactions on Industrial Electronics, 2026, 73(6): 9174−9186 doi: 10.1109/TIE.2025.3649849
[6]	Tian G Y, Yang Y, Wen S P. Time-series stock price forecasting based on neural networks: A comprehensive survey. Artificial Intelligence Science and Engineering, 2025, 1(4): 255−277 doi: 10.23919/AISE.2025.000018
[7]	肖敏, 陆云翔, 虞文武, 郑卫新. 神经网络分岔动力学综述. 自动化学报, 2025, 51(1): 72−89 doi: 10.16383/j.aas.c230789 Xiao Min, Lu Yun-Xiang, Yu Wen-Wu, Zheng Wei-Xin. Overview of bifurcation dynamics in neural networks. Acta Automatica Sinica, 2025, 51(1): 72−89 doi: 10.16383/j.aas.c230789
[8]	Zhang J, Zhu S, Wu K N, Shen M Q, Wen S P. Finite-time stabilization of semi-Markov reaction-diffusion memristive NNs with unbounded time-varying delay. IEEE Transactions on Circuits and Systems I: Regular Papers, 2025, 72(4): 1832−1842 doi: 10.1109/TCSI.2024.3459913
[9]	Zhao S S, Zhao L H, Wen S. Secure synchronization control of Markovian jump neural networks under DoS attacks with memory-based adaptive event-triggered mechanism. Artificial Intelligence Science and Engineering, 2025, 1(1): 64−78 doi: 10.23919/AISE.2025.000006
[10]	Wang L M, Zhang C K. Exponential synchronization of memristor-based competitive neural networks with reaction-diffusions and infinite distributed delays. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(1): 745−758 doi: 10.1109/TNNLS.2022.3176887
[11]	Wen G H, Yu W W, Hu G Q, Cao J D, Yu X H. Pinning synchronization of directed networks with switching topologies: A multiple Lyapunov functions approach. IEEE Transactions on Neural Networks and Learning Systems, 2015, 26(12): 3239−3250 doi: 10.1109/TNNLS.2015.2443064
[12]	Wang P J, Yu W W, Huang T W, Lv Y Z. Consensus disturbance rejection of mimo linear multiagent systems with directed switching topologies: A reduced order observer approach. IEEE Transactions on Circuits and Systems II: Express Briefs, 2022, 69(12): 4984−4988
[13]	孙亚平, 杨鑫松, 苏厚胜. 随机切换和不确定目标下异质非线性多智能体系统的一致性跟踪控制. 中国科学: 信息科学, 2024, 54: 2674−2688 Sun Y P, Yang X S, Su H S. Consensus tracking control of heterogeneous nonlinear multi-agent systems with random switching and uncertain targets (in Chinese). Sci Sin Inform, 2024, 54: 2674−2688
[14]	Li B Q, Zhao L H, Wen S P. Periodic event-triggered consensus of stochastic multi-agent systems under switching topology. Artificial Intelligence Science and Engineering, 2025, 1(2): 147−156 doi: 10.23919/AISE.2025.000011
[15]	Cao Y T, Zhao L H, Zhong Q S, Shi K B, Xiao J Y, Wen S P. Synchronization of coupled neural networks with multiple switching topologies via adaptive control techniques. International Journal of Robust and Nonlinear Control, 2024, 34(13): 8661−8675 doi: 10.1002/rnc.7409
[16]	Wang S, Cao Y T, Guo Z Y, Yan Z, Wen S P, Huang T W. Periodic event-triggered synchronization of multiple memristive neural networks with switching topologies and parameter mismatch. IEEE Transactions on Cybernetics, 2021, 51(1): 427−437 doi: 10.1109/TCYB.2020.2983481
[17]	Chen J, Chen B, Zeng Z. Synchronization in multiple neural networks with delay and disconnected switching topology via event-triggered impulsive control strategy. IEEE Transactions on Industrial Electronics, 2021, 68(3): 2491−2500 doi: 10.1109/TIE.2020.2975498
[18]	Cao Y Y, Liu N, Zhang C F, et al. Synchronization of multiple reaction–diffusion memristive neural networks with known or unknown parameters and switching topologies. Knowledge-Based Systems, 2022, 254: Article No. 109595 doi: 10.1016/j.knosys.2022.109595
[19]	Miao B J, Li X C, Lou J G, Lu J Q. Pinning bipartite synchronization for coupled reaction–diffusion neural networks with antagonistic interactions and switching topologies. Neural Networks, 2021, 141: 174−183 doi: 10.1016/j.neunet.2021.04.007
[20]	Sun Y P, Yang X S, Su H S. Fully distributed observer-based scaled consensus of multi-agent systems with actuator saturation and edge-based event-triggered communication. Neurocomputing, 2024, 600: Article No. 128134 doi: 10.1016/j.neucom.2024.128134
[21]	Wu T, Cao J D, Xiong L L, Park J H, Tan C G. Adaptive event-triggered mechanism to synchronization of reaction–diffusion CVNNs and its application in image secure communication. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2023, 53(9): 5307−5320 doi: 10.1109/TSMC.2023.3258222
[22]	Qiu Q, Su H. Sampling-based event-triggered exponential synchronization for reaction-diffusion neural network. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(3): 1209−1217 doi: 10.1109/TNNLS.2021.3105126
[23]	Ge F, Chen Y Q. Observer-based boundary stabilization of coupled semilinear reaction–diffusion neural networks with spatially varying coefficients via event-triggered controller. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(6): 8348−8357 doi: 10.1109/TNNLS.2022.3227109
[24]	Cao Y Y, Cao Y T. Synchronization of multiple neural networks with reaction–diffusion terms under cyber–physical attacks. Knowledge-Based Systems, 2022, 239: Article No. 107939 doi: 10.1016/j.knosys.2021.107939
[25]	Zhang R M, Wang H X, Park J H, Lam H K, He P S. Quasisynchronization of reaction–diffusion neural networks under deception attacks. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2022, 52(12): 7833−7844 doi: 10.1109/TSMC.2022.3166554
[26]	Cao Y, Cao Y. Event-triggered synchronization for delayed reaction–diffusion neural networks under hybrid deception attacks. Knowledge-Based Systems, 2024, 301: Article No. 112304 doi: 10.1016/j.knosys.2024.112304
[27]	Wang Z P, Li Q Q, Wu H N, et al. Pinning spatiotemporal sampled-data synchronization of coupled reaction–diffusion neural networks under deception attacks. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(10): 7967−7977 doi: 10.1109/TNNLS.2022.3148184
[28]	Zhang H, Zeng Z G. Synchronization of nonidentical neural networks with unknown parameters and diffusion effects via robust adaptive control techniques. IEEE Transactions on Cybernetics, 2021, 51(2): 660−672 doi: 10.1109/TCYB.2019.2921633