Decentralized Learning With Limited Communications for Multi-robot Coverage of Unknown Spatial Fields

TL;DR

This paper introduces a decentralized algorithm for multi-robot teams to learn and cover unknown spatial fields efficiently with limited communication, using local Gaussian processes and Voronoi-based information sharing.

Contribution

It proposes a novel decentralized approach combining local Gaussian processes and Voronoi communication boundaries for efficient spatial learning and coverage.

Findings

Algorithm performs comparably to centralized methods in simulations.

Limited communication reduces data exchange while maintaining accuracy.

Decentralized strategy scales with number of robots and domain size.

Abstract

This paper presents an algorithm for a team of mobile robots to simultaneously learn a spatial field over a domain and spatially distribute themselves to optimally cover it. Drawing from previous approaches that estimate the spatial field through a centralized Gaussian process, this work leverages the spatial structure of the coverage problem and presents a decentralized strategy where samples are aggregated locally by establishing communications through the boundaries of a Voronoi partition. We present an algorithm whereby each robot runs a local Gaussian process calculated from its own measurements and those provided by its Voronoi neighbors, which are incorporated into the individual robot's Gaussian process only if they provide sufficiently novel information. The performance of the algorithm is evaluated in simulation and compared with centralized approaches.