Optimal Network Topology of Multi-Agent Systems subject to Computation and Communication Latency (with proofs)

TL;DR

This paper develops an efficient method for designing minimum-variance feedback controllers in multi-agent systems with communication delays, optimizing network topology for improved control performance.

Contribution

It introduces a closed-form solution for near-optimal control gains in circular formations and proposes a control-driven network topology optimization.

Findings

Closed-form near-optimal control gains for circular formations.

Centralized control is suboptimal as network size grows.

Network topology optimization improves control performance.

Abstract

We study minimum-variance feedback-control design for a networked control system with retarded dynamics, where inter-agent communication is subject to latency. We prove that such a design can be solved efficiently for circular formations and compute near-optimal control gains in closed form. We show that the centralized control architecture is in general suboptimal when the communication increase with the number of links, and propose a control-driven optimization of the network topology.