Periodic Event-Triggered SMC for Unknown Linear Systems Under Dual-Channel FDI Attacks: A Data-Based Method.

This work considers the problem of event-triggered sliding mode secure control for unknown linear systems, in which the transmission of sampled states is regulated via a periodic event-triggered mechanism (ETM); meanwhile, the false data injection (FDI) attack may occur in the dual channels (S/C and C/A channels). First, an event-triggered sliding mode controller is designed, and the model-based sufficient conditions are derived to guarantee the boundedness of the closed-loop system and the reachability of the sliding surface. Since the actual system parameters may not be completely known/available for the controller, this work shall further focus on the following cases: the system parameters are completely unknown or partially unknown. By introducing the data-based representations for various cases (i.e., the system parameters are completely unknown and partially unknown), the above model-based sufficient conditions are transformed into the data-based forms or the combination forms of data and partially known parameters, under which the designed secure controller and ETM can be implemented via system data and partially known system parameters. Compared with traditional model-based methods, the core advantage of the proposed data-based control scheme lies in completely eliminating the reliance on precise mathematical models and designing the controller solely using offline collected input/state data and partially available parameters. Finally, simulation results are provided to illustrate the proposed data-based control scheme.