Neural-network-based output feedback adaptive resilient control for stochastic nonlinear systems subject to sensor and actuator attacks

Abstract

This paper addresses the output feedback adaptive resilient control problem for a class of stochastic nonlinear systems under sensor and actuator attacks. By applying a linear state transformation, the original system is transformed into a new one for which output feedback control becomes possible. The Nussbaum function is exploited to overcome the unknown control direction problem generated by uncertain control coefficients and adversary attacks. Based on the available compromised output signal and the states reconstructed by a new extended state observer, an output feedback adaptive resilient control scheme is proposed for such systems by combining the neural-networks-based approximation approach, backstepping technique and some special methods. The proposed controller can guarantee that all signals in the closed-loop system are semiglobally bounded in probability and the stabilisation error variables remain in an adjustable vicinity of origin. Finally, the effectiveness and applicability of the proposed control methodology is verified by two illustrative examples.

Keywords:

• Stochastic nonlinear system
• sensor and actuator attacks
• adaptive resilient control
• output feedback
• Nussbaum gain function

Acknowledgments

The authors would like to appreciate the editors and reviewers for their valuable and constructive comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work is partially supported by the Chinese National Natural Science Foundation [grant 71871135] and Fundamental Research Funds for the Central Universities [grant 222201717006].