Observer-based dynamic event-triggered second-level MPC for nonlinear time-delay CPSs under joint hybrid attacks

Abstract

This study investigates observer-based dynamic event-triggered (DET) second-level model predictive control (MPC) for a specific class of nonlinear time-delay cyber-physical systems (CPSs) under hybrid attacks on the sensor-controller (S-C) channel and controller-actuator (C-A) channel. A dynamic event-triggered mechanism with adaptive bias components is developed to reduce trigger frequency, uphold system performance, and conserve communication resources. Additionally, the observer gain is determined using the nominal observation model and linear matrix inequalities. A novel model predictive control approach is devised based on the identified model to enhance system performance. Theoretical analysis demonstrates that the proposed method effectively guarantees that the closed-loop system satisfies the input-to-state practical stability criterion. Finally, a numerical simulation case study is presented to validate the effectiveness of the proposed algorithm.