Algebraic Connectivity Control in Distributed Networks by Using Multiple Communication Channels

Area of Science:

• Distributed systems
• Network science
• Control theory

Background:

• Effective collaboration in distributed networks hinges on reliable information exchange between nodes.
• Network connectivity, crucial for system operation, can be quantified using algebraic connectivity (Fiedler value).
• Existing methods may not adequately address dynamic network conditions and diverse communication protocols.

Purpose of the Study:

• To implement a trust-based consensus algorithm for estimating algebraic connectivity in distributed systems.
• To develop a decentralized framework for combining multiple communication channels.
• To design a decentralized control law for dynamic adjustment of agent numbers to maintain connectivity.

Main Methods:

• A trust-based consensus algorithm was developed for decentralized algebraic connectivity estimation.
• A novel framework was created to merge diverse communication channels into a unified channel.
• A decentralized control law was implemented to dynamically manage the number of agents.

Main Results:

• The proposed trust-based algorithm effectively estimates algebraic connectivity in distributed networks.
• The decentralized framework successfully combined channels with different properties.
• The control strategy dynamically adjusted agent numbers, ensuring system connectivity during operation.
• Simulations and real-world multi-robot experiments validated the approach.

Conclusions:

• The developed trust-based consensus algorithm and decentralized control strategy enhance connectivity in distributed systems.
• The framework offers a robust solution for managing communication channels and agent numbers in dynamic environments.
• This approach improves the reliability and performance of multi-agent and sensor networks.