Event-Based Adaptive Control for MIMO Nonlinear Systems Against Deception Attacks via a Single Parameter Learning Method

This article proposes an adaptive neural network (NN) control scheme for MIMO nonlinear systems with unknown deception attacks. Based on the dynamic surface technology and the backstepping method, the unknown term composed of the sensor attack signals, the error term generated by NN and the disturbance can be regarded as an uncertain parameter. In addition, by combining the adaptive technology with the single parameter learning method (SPLM), the uncertain parameter will be transformed into a linear parameterized form with only one unknown uncertain parameter, thereby significantly reducing the computational complexity in the recursive process. Furthermore, the event-triggered control strategy with dynamic characteristics can be integrated into the control design to reduce the transmission of data. Theoretical analysis demonstrates that the proposed control algorithm ensures boundedness of all signals in the closed-loop system, and the Zeno phenomenon is eliminated. Ultimately, a numerical simulation is employed to validate the effectiveness of the proposed NN adaptive control scheme.