Preset-Time Tracking Control for Nonlinear Cyber-Physical Systems Against Cyber Attacks

In this article, we developed a novel preset-time adaptive fuzzy tracking control strategy to dispose of the preset-time tracking control issue for a family of nonlinear cyber-physical systems (CPSs), which are subject to spiteful actuator attacks occurring in controller-actuator (C-A) channel. Particularly, a new Gaussian radial basis function neural network (RBFNN) is designed as a neural estimator acting in an on-line pattern to estimate the vitriolic attacks. The designed control strategy can guarantee that the output tracking error enters a predefined small region around the equilibrium point, but also all the other states of the closed-loop system keep semi-globally practical finite-time stable (SGPFS), where the regulation time and tracking accuracy level remain prior known and could be preassigned. Significantly, the initial value of each system state is independent of the developed tracking controller and can be chosen in the constrained region specified by the designed finite-time performance function at random. Finally, a typical application instance is presented to attest the feasibility and effectiveness of the designed control approach.