Resilient Adaptive Stabilization for Nonlinear CPSs Dealing With Deception Attacks
Xinjun Wang, Ye Cao, Ben Niu, Huanqing Wang, Guangdeng Zong, Xudong Zhao
Abstract
It is still an open problem to output feedback stabilization with input quantizations and deception attacks for nonlinear cyber-physical systems subject to mismatched parametric uncertainties via backstepping design. This is because: first, the system contains mismatched uncertainties; second the output is corrupted by an additional attack signal, making the conventional recursive control design strategy using traditional error surfaces unsuitable; and third, since only the compromised output signal is available, the impacts of nonlinear items related to unknown attack weights become difficult to be explicitly addressed in the existing available output feedback control results. In this article, we present a solution to circumvent these obstacles by constructing a set of new <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$K$</tex-math></inline-formula>-filters using compromised output only, which, together with a coordinate transformation involving the attacked output and filter variables, allow a new form of state-observer-based resilient adaptive control scheme to be developed. Meanwhile, a novel compensation term is incorporated into the real controller to eliminate the effect of quantization on the system stability. It is shown that, by establishing Lyapunov functions appropriately, the original output of the system converges to zero, and all the signals in the closed-loop system are globally uniformly bounded. A practice example is provided to confirm the effectiveness of the proposed strategy.