Adaptive Resilient Output Feedback Control Against Unknown Deception Attacks for Nonlinear Cyber-Physical Systems

In this paper, by considering the sensors subject to deception attacks, the observer-based adaptive resilient control problem of a class of nonlinear cyber-physical systems is addressed. Because authentic system state information can be injected with counterfeit data, the impacts of unknown nonlinearity related to unknown attack weights become difficult to be compensated. To circumvent these obstacles, we construct a novel state observer using the system output after being attacked only, and the effects of unknown control gains due to deception attacks on the system are handled by introducing Nussbaum technique. A resilient output feedback control scheme is designed by utilizing the compromised state variables, guaranteeing boundedness for all signals within the closed-loop system. In the end, the efficacy of the proposed control scheme is affirmed through simulation results involving a single-link robot.