Design of Finite‐Time Adaptive Controller for Nonlinear Cyber‐Physical Systems Under Deception Attacks via Event‐Triggered Strategy

ABSTRACT In this paper, the problem of finite‐time adaptive neural network (NN) event‐triggered (ET) control is investigated for a class of nonlinear cyber‐physical systems with deception attacks by a command filtered technique. To achieve better noise reduction property and faster convergence, a second‐order command filter is introduced and the finite‐time control problem is considered. Since the use of the filter requires the second‐order derivative of virtual controller to be existent, it cannot be directly applied to the finite‐time control design and therefore, a switching‐like function with second‐order derivative is designed to overcome this obstacle. On the other hand, due to the existence of unknown deception attacks, only partial compromised information is available, which brings difficulty to the design of finite‐time ET control. Based on the command filtered and adaptive NN techniques, an adaptive finite‐time ET controller is designed to guarantee that all states of the closed‐loop systems can converge to a compact set in a finite time. Finally, the Zeno behavior is successfully excluded and a two‐stage chemical reactor system is provided to demonstrate the effectiveness and merits of the proposed method.