Improving the consensus performance via predictive mechanisms

TL;DR

This paper demonstrates that incorporating predictive mechanisms into consensus protocols significantly accelerates convergence and broadens operational parameters, with implications for biological systems and industrial applications.

Contribution

It introduces a novel predictive approach to consensus protocols that enhances speed and robustness without altering network topology.

Findings

Ultrafast consensus achieved through prediction-based protocols

Expanded sampling periods for reliable consensus

Reduced communication energy in network consensus

Abstract

Considering some predictive mechanisms, we show that ultrafast average-consensus can be achieved in networks of interconnected agents. More specifically, by predicting the dynamics of the network several steps ahead and using this information in the design of the consensus protocol of each agent, drastic improvements can be achieved in terms of the speed of consensus convergence, without changing the topology of the network. Moreover, using these predictive mechanisms, the range of sampling periods leading to consensus convergence is greatly expanded compared with the routine consensus protocol. This study provides a mathematical basis for the idea that some predictive mechanisms exist in widely-spread biological swarms, flocks, and networks. From the industrial engineering point of view, inclusion of an efficient predictive mechanism allows for a significant increase in the speed of consensus convergence and also a reduction of the communication energy required to achieve a predefined consensus performance.