Federated Gaussian Process Learning via Pseudo-Representations for Large-Scale Multi-Robot Systems

TL;DR

This paper presents pxpGP, a scalable federated Gaussian Process framework for large multi-robot systems, using pseudo-representations and advanced optimization to improve efficiency and accuracy in environment modeling.

Contribution

Introduces pxpGP, a novel distributed Gaussian Process method utilizing sparse variational inference and ADMM for large-scale multi-robot systems, enhancing scalability and performance.

Findings

Outperforms existing distributed GP methods in accuracy and hyperparameter estimation.

Effective in both centralized and decentralized large-scale networks.

Demonstrates robustness on synthetic and real-world datasets.

Abstract

Multi-robot systems require scalable and federated methods to model complex environments under computational and communication constraints. Gaussian Processes (GPs) offer robust probabilistic modeling, but suffer from cubic computational complexity, limiting their applicability in large-scale deployments. To address this challenge, we introduce the pxpGP, a novel distributed GP framework tailored for both centralized and decentralized large-scale multi-robot networks. Our approach leverages sparse variational inference to generate a local compact pseudo-representation. We introduce a sparse variational optimization scheme that bounds local pseudo-datasets and formulate a global scaled proximal-inexact consensus alternating direction method of multipliers (ADMM) with adaptive parameter updates and warm-start initialization. Experiments on synthetic and real-world datasets demonstrate that pxpGP and its decentralized variant, dec-pxpGP, outperform existing distributed GP methods in hyperparameter estimation and prediction accuracy, particularly in large-scale networks.