Adaptive Resilient Neural Control of Uncertain Time-Delay Nonlinear CPSs with Full-State Constraints under Deception...

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems Security
• Nonlinear Systems Theory

Background:

• Cyber-physical systems (CPSs) are vulnerable to unknown time-varying deception attacks and state constraints.
• Sensor disturbances from attacks obscure system state variables, complicating control design.
• Traditional control methods may struggle with computational complexity and unknown system dynamics.

Purpose of the Study:

• To develop an adaptive resilient control strategy for uncertain time-delay nonlinear CPSs.
• To address challenges posed by unknown deception attacks and full-state constraints.
• To ensure system stability and performance under adversarial conditions.

Main Methods:

• A novel backstepping control strategy using compromised variables and dynamic surface techniques.
• Introduction of attack compensators to counteract unknown attack signals.
• Utilization of barrier Lyapunov functions (BLF) for state variable constraints.
• Approximation of unknown nonlinear terms using radial basis function (RBF) neural networks.
• Application of Lyapunov-Krasovskii functions (LKF) to handle unknown time-delay terms.

Main Results:

• Design of an adaptive resilient controller guaranteeing convergence of system state variables.
• Ensured satisfaction of predefined state constraints for the CPS.
• Demonstration of semi-globally uniformly ultimately boundedness for all closed-loop system signals.
• Validation of theoretical results through numerical simulation experiments.

Conclusions:

• The proposed adaptive resilient control strategy effectively manages uncertain time-delay nonlinear CPSs under deception attacks and state constraints.
• The controller ensures both system stability and adherence to state limitations.
• The combination of advanced control techniques provides a robust solution for secure CPS operation.