A Unified Optimization for Resilient Dynamic Event-Triggering Consensus Under Denial of Service

Area of Science:

• Control Systems Engineering
• Network Security
• Robotics

Background:

• Multiagent systems (MAS) are vulnerable to denial-of-service (DoS) attacks that disrupt communication.
• Existing consensus protocols may not guarantee resilience against such adversarial network conditions.

Purpose of the Study:

• To propose a resilient framework for optimized consensus in MAS under DoS attacks.
• To develop a dynamic event-triggering (DET) scheme for enhanced communication resilience.
• To co-design control and DET parameters for improved system performance.

Main Methods:

• A distributed dynamic event-triggering (DET) scheme to limit inter-agent transmissions.
• A novel convex optimization approach to co-design control and DET parameters.
• Weighted sum approach for optimization, increasing inter-event intervals.
• Incorporating desired resilience level as an input to the co-design framework.

Main Results:

• The proposed framework achieves a desired level of resilience to DoS attacks.
• The co-design approach provides a less conservative (larger) upper bound for guaranteed resilience.
• A structured tradeoff is established between consensus convergence rate, event triggering frequency, and DoS resilience.
• Simulations with nonholonomic mobile robots demonstrate the framework's effectiveness.

Conclusions:

• The proposed resilient framework effectively optimizes consensus in MAS under DoS attacks.
• The co-design approach offers a significant improvement in guaranteed resilience compared to existing methods.
• The framework provides a valuable tool for designing robust and secure multiagent systems.