欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2025, Vol. 16 ›› Issue (6): 914-922.DOI: 10.3969/j.issn.1674-8484.2025.06.011

• 智能驾驶与智慧交通 • 上一篇    下一篇

隐蔽网络攻击下智能网联汽车容错安全控制

邱照玉1(), 祝小元1,*(), 田光宇2, 殷国栋1   

  1. 1.东南大学 机械工程学院，南京 211189，中国
    2.清华大学 车辆与运载学院，北京 100084，中国
  • 收稿日期:2025-06-05 修回日期:2025-11-18 出版日期:2025-12-31 发布日期:2026-01-12
  • 通讯作者: * 祝小元, 副教授。E-mail：zhuxy@seu.edu.cn。
  • 作者简介:邱照玉（1996—），男（汉），山东, 博士研究生。E-mail：qiuzhaoyu@seu.edu.cn。
  • 基金资助:
    智能绿色车辆与交通全国重点实验室开放基金课题(KFZ2406);国家自然科学基金项目(52172402);国家自然科学基金项目(52394264);东南大学博士生创新能力提升计划(CXJH_SEU 24056)

Fault-tolerant and safety control of intelligent connected vehicles under stealthy network attacks

QIU Zhaoyu1(), ZHU Xiaoyuan1,*(), TIAN Guangyu2, YIN Guodong1   

  1. 1. School of Mechanical and Automotive Engineering, Xiamen Institute of Technology, Xiamen 361024, China
    2. Fujian Key Laboratory of Advanced Design and Manufacturing of Buses, Xiamen 361024, China
  • Received:2025-06-05 Revised:2025-11-18 Online:2025-12-31 Published:2026-01-12

RichHTML

6

PDF

72

摘要： 针对智能网联汽车面临的执行器故障和隐蔽重放攻击双重安全威胁，为提高车辆安全性，该文提出一种融合动态水印攻击检测技术的自适应神经网络控制方法。通过集成径向基神经网络（RBFNN）与非线性扰动观测器（NDO），设计了具有扰动抑制能力的自适应容错控制器。此外，在控制回路中嵌入动态水印信号，基于系统残差构建攻击检测机制，用于检测隐蔽重放攻击。最后，利用dSPACE-NI联合仿真平台进行硬件在环（HIL）验证。结果表明：该方法可有效降低故障对控制系统的影响；故障期间，自适应容错控制器相较于无容错的控制器，平均误差降低了80.71%，并能成功检测出系统中的隐蔽性重放攻击，同时故障的存在增强了检测效果，且不会导致攻击检测误报。

关键词: 径向基神经网络（RBFNN）, 容错控制, 隐蔽性重放攻击检测, 动态水印

Abstract:

An adaptive neural network control method integrated with dynamic watermark-based attack detection to enhance vehicle safety was proposed to address the dual safety threats of actuator faults and stealthy replay attacks in intelligent connected vehicles. An adaptive fault-tolerant controller with disturbance rejection capability was designed by integrating a radial basis function neural network (RBFNN) and a nonlinear disturbance observer (NDO). Additionally, a dynamic watermark sequence was embedded into the control loop, and an attack detection mechanism was constructed based on system residuals to identify covert replay network attacks. Finally, hardware-in-the-loop (HIL) validation was conducted using a dSPACE-NI co-simulation platform. The results show that the average error during the fault is reduced by 80.71%, comparing with the non-fault-tolerant controller. Furthermore, stealthy replay attacks are successfully detected, and the presence of faults enhances the detection effectiveness without causing false alarms.

Key words: radial basis function neural network (RBFNN), fault-tolerant control, stealthy replay attack detection, dynamic watermarking

中图分类号: