Control of Multiagent Formations

TL;DR

This paper introduces a distributed control architecture for multiagent formations that leverages multiplex information networks to enable agents to modify formation density and orientation without global information.

Contribution

It presents a novel approach to design information exchange rules in multiplex networks for controlling multiagent formations.

Findings

Distributed control architecture enables formation adjustments.

Agents can alter formation density and orientation locally.

No need for global information exchange.

Abstract

This paper makes the first attempt to show how information exchange rules represented by a network having multiple layers (multiplex information networks) can be designed for enabling spatially evolving multiagent formations. Toward this goal, we consider the invariant formation problem and introduce a distributed control architecture that allows capable agents to spatially alter the resulting formation's density and orientation without requiring global information exchange ability.