An adaptive predefined time sliding mode control for uncertain nonlinear cyber-physical servo system under cyber...

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems (CPS)
• Robotics and Automation

Background:

• Cyber-physical systems (CPS) face inevitable malicious attacks, impacting critical functions like servo position tracking.
• Achieving high accuracy in industrial automation is challenging due to cyber-attacks, control saturations, parametric perturbations, and external disturbances.
• Existing control methods often struggle with robustness and singularity issues in complex CPS environments.

Purpose of the Study:

• To design a novel predefined-time convergence sliding mode adaptive controller (PTCSMAC) for robust servo tracking in CPS.
• To address challenges including malicious cyber-attacks, control saturation, and parameter uncertainties.
• To enhance system performance and eliminate singularity problems in CPS control.

Main Methods:

• Designed a novel predefined-time (PDT) convergence sliding mode adaptive controller (PTCSMAC).
• Upgraded the system to a third-order system to facilitate an adaptive control law.
• Integrated a modified weight-updated Extreme Learning Machine (ELM) to approximate system uncertainties.
• Ensured the controller is nonsingular, irrespective of initial conditions.

Main Results:

• The proposed PTCSMAC effectively copes with parameter perturbation, control saturation, and cyber-attacks.
• The controller demonstrates robust performance even with uncertain system models, without needing detailed model information.
• Rigorous simulations on a CPS seeker servo positioning system verified the control law's effectiveness and robustness.
• The PTCSMAC successfully eradicated singularity problems common in CPS control.

Conclusions:

• The developed PTCSMAC offers a robust and accurate solution for servo position tracking in CPS under adversarial conditions.
• The controller's model-free nature and nonsingular design provide significant advantages for practical CPS applications.
• The study highlights the potential of PDT convergence and ELM integration for enhancing CPS security and performance.