Leader-following consensus for multi-agent systems subject to cyber attacks: Dynamic event-triggered control.

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems
• Networked Systems

Background:

• Multi-agent systems (MASs) face challenges in achieving consensus, especially when subjected to cyber attacks.
• Existing event-triggered mechanisms may not adequately address the complexities of recoverable cyber attacks and Zeno phenomenon.

Purpose of the Study:

• To address the event-triggered consensus problem for MASs under recoverable cyber attacks.
• To co-design centralized and distributed event-triggered controllers using a dynamic event-triggered mechanism (DEM).
• To ensure MASs achieve consensus despite cyber threats and avoid Zeno phenomenon.

Main Methods:

• Introduction of a recoverable cyber attack model.
• Development and implementation of a dynamic event-triggered mechanism (DEM) with time-varying parameters.
• Co-design of centralized and distributed event-triggered controllers.
• Formulation of dynamic triggering criteria for consensus achievement.

Main Results:

• Successful co-design of controllers for leader-following MASs.
• Demonstration of consensus achievement under cyber attacks.
• Extension of interevent time intervals via DEM, effectively avoiding the Zeno phenomenon.
• Validation of the proposed scheme's effectiveness through simulation.

Conclusions:

• The proposed dynamic event-triggered mechanism and co-designed controllers effectively achieve consensus in MASs under cyber attacks.
• The DEM provides a robust solution for managing communication and control updates, enhancing system resilience.
• The approach successfully mitigates the Zeno phenomenon, ensuring practical implementation of event-triggered control.