Event-triggered distributed optimization of multi-agent systems with time delay

Area of Science:

• Control Systems
• Optimization Theory
• Distributed Computing

Background:

• Distributed optimization is crucial for multi-agent systems.
• Continuous communication in these systems can be inefficient.
• Existing methods may not adequately address time delays or communication constraints.

Purpose of the Study:

• To develop an event-triggered distributed optimization algorithm for multi-agent systems.
• To reduce communication load by triggering updates only when necessary.
• To analyze the algorithm's performance under time-delayed conditions.

Main Methods:

• A novel event-triggering mechanism based on state measurement error and decay thresholds was designed.
• The algorithm incorporates considerations for time delays in agent communication.
• Sufficient conditions for achieving optimal consistency were mathematically derived.

Main Results:

• The proposed event-triggered algorithm effectively reduces inter-agent communication.
• Optimal consistency was achieved under defined conditions, despite time delays.
• The algorithm successfully avoids Zeno behavior, ensuring practical implementation.

Conclusions:

• The developed event-triggered distributed optimization algorithm is effective for multi-agent systems.
• This approach offers a more efficient communication strategy compared to traditional methods.
• The algorithm provides a robust solution for optimizing systems with time delays.