Distributed interval state estimation with l_∞-gain optimization for cyber-physical systems subject to bounded...

Area of Science:

• Cyber-physical systems security
• Control theory and applications
• Distributed state estimation

Background:

• Conventional interval observers struggle with real-time monitoring under random stealthy attacks in cyber-physical systems.
• Bounded disturbances and random stealthy attacks pose significant challenges to system state estimation.
• Need for robust observers capable of handling uncertainties and adversarial conditions.

Purpose of the Study:

• To design an attack-resistant distributed interval observer for cyber-physical systems.
• To ensure reliable real-time monitoring despite bounded disturbances and random stealthy attacks.
• To develop methods for estimating state intervals under adversarial conditions.

Main Methods:

• Design of an attack-resistant distributed interval observer utilizing attack frequency and interval attack estimation.
• Modeling of upper- and lower-bounding estimation error systems as positive interconnected systems.
• Formulation of a deterministic positive error system to attenuate disturbance and attack effects.
• Application of linear programming for interval observer design and L∞-gain optimization.

Main Results:

• Successful design of an attack-resistant distributed interval observer.
• Demonstration of effective attenuation of disturbance and attack effects on estimation errors.
• Validation of the observer's performance through remote monitoring of vehicle lateral dynamics.
• Achievement of interval observer design and L∞-gain optimization results via linear programming.

Conclusions:

• The proposed attack-resistant distributed interval observer effectively addresses challenges in cyber-physical systems.
• The method ensures robust real-time monitoring and state estimation under complex disturbances and attacks.
• Numerical verification confirms the practical applicability and performance of the developed observer.