Scholarlink

<p>This paper is concerned with the secure control problem for Takagi–Sugeno fuzzy systems under sensor and actuator attacks. Since attackers can simultaneously compromise the sensor measurement and control input, the premise variables of the fuzzy controller, which depend on the measurable states, and the actual control signal become imprecise, potentially leaving the traditional control strategies invalid. To address this issue, the actual sensor measurements are recovered from a set of corrupted ones via robust principal component analysis to further construct accurate premise variables of the controller. Moreover, by leveraging the upper bound information on the 2-norm of malicious attacks, a novel fuzzy controller with an adaptive compensator against actuator attacks is proposed, where a design condition in terms of linear matrix inequalities is also given, to guarantee the asymptotic stability of the attacked system for the disturbance-free case and the desired <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\({\mathcal {H}}_\infty \)</EquationSource> <EquationSource Format="MATHML"><math> <msub> <mi mathvariant="script">H</mi> <mi>∞</mi> </msub> </math></EquationSource> </InlineEquation> performance under disturbance. Finally, simulation results are provided to verify the effectiveness and advantages of the control strategy.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive Secure Control for T–S Fuzzy Systems Against Sensor and Actuator Attacks Based on Robust Principal Component Analysis

  • 期刊论文

  • 发表日期：

    2026年3月9日

查看全文
  • Yongbao Xiao,
  • Xiaojian Li

摘要

This paper is concerned with the secure control problem for Takagi–Sugeno fuzzy systems under sensor and actuator attacks. Since attackers can simultaneously compromise the sensor measurement and control input, the premise variables of the fuzzy controller, which depend on the measurable states, and the actual control signal become imprecise, potentially leaving the traditional control strategies invalid. To address this issue, the actual sensor measurements are recovered from a set of corrupted ones via robust principal component analysis to further construct accurate premise variables of the controller. Moreover, by leveraging the upper bound information on the 2-norm of malicious attacks, a novel fuzzy controller with an adaptive compensator against actuator attacks is proposed, where a design condition in terms of linear matrix inequalities is also given, to guarantee the asymptotic stability of the attacked system for the disturbance-free case and the desired \({\mathcal {H}}_\infty \) H ∞ performance under disturbance. Finally, simulation results are provided to verify the effectiveness and advantages of the control strategy.