Secure predictor-based neural dynamic surface control of nonlinear cyber–physical systems against sensor and actuator attacks.

This paper addresses a secure predictor-based neural dynamic surface control (SPNDSC) issue for a cyber–physical system in a nontriangular form suffering from both sensor and actuator deception attacks. To avoid the algebraic loop problem, only partial states are employed as input vectors of neural networks (NNs) for approximating unknown dynamics, and compensation terms are further developed to offset approximation errors from NNs. With introduction of nonlinear gain functions and attack compensators, adverse effects of an intelligent adversary are alleviated effectively. Furthermore, we present stability analysis and prove the ultimate boundedness of all signals in the closed-loop system. The effectiveness of the proposed control strategy is illustrated by two examples. • A secure predictor-based neural dynamic surface control (SPNDSC) strategy is proposed to accommodate the compromised state information. • By introducing a nonlinear gain function into our controller, the negative effects caused by an intelligent adversary are mitigated. • We utilize NNs to estimate the upper bound of the actuator attacks and compensate their adverse impacts, and design sensor attack compensators to deal with the sensor deception attacks. • An improved predictor is developed to achieve NNs' learning performance, while overcoming the so-called algebraic loop issue without strict assumptions. [ABSTRACT FROM AUTHOR]