Theorems about Ergodicity and Class-Ergodicity of Chains with Applications in Known Consensus Models

TL;DR

This paper establishes conditions for ergodicity and class-ergodicity in chains related to multi-agent consensus models, providing insights into when agents reach agreement regardless of initial states.

Contribution

It introduces necessary and sufficient conditions for ergodicity in chains with balanced asymmetry, applying these to analyze consensus in several well-known models.

Findings

Unconditional consensus occurs in JLM, Krause, and Cucker-Smale models.

Necessary and sufficient conditions for consensus in JLM model.

Sufficient conditions for consensus in Cucker-Smale model.

Abstract

In a multi-agent system, unconditional (multiple) consensus is the property of reaching to (multiple) consensus irrespective of the instant and values at which states are initialized. For linear algorithms, occurrence of unconditional (multiple) consensus turns out to be equivalent to (class-) ergodicity of the transition chain (A_n). For a wide class of chains, chains with so-called balanced asymmetry property, necessary and sufficient conditions for ergodicity and class-ergodicity are derived. The results are employed to analyze the limiting behavior of agents' states in the JLM model, the Krause model, and the Cucker-Smale model. In particular, unconditional single or multiple consensus occurs in all three models. Moreover, a necessary and sufficient condition for unconditional consensus in the JLM model and a sufficient condition for consensus in the Cucker-Smale model are obtained.