Fleet Control using Coregionalized Gaussian Process Policy Iteration

Onderzoeksoutput: Conference paper

2 Citaten (Scopus)
47 Downloads (Pure)

Samenvatting

In many settings, as for example wind farms, multiple machines are instantiated to perform the same task, which is called a fleet. The recent advances with respect to the Internet of Things allow control devices and/or machines to connect through cloud-based architectures in order to share information about their status and environment. Such an infrastructure allows seamless data sharing between fleet members, which could greatly improve the sample-efficiency of reinforcement learning techniques. However in practice, these machines, while almost identical in design, have small discrepancies due to production errors or degradation, preventing control algorithms to simply aggregate and employ all fleet data. We propose a novel reinforcement learning method that learns to transfer knowledge between similar fleet members and creates member-specific dynamical models for control. Our algorithm uses Gaussian processes to establish cross-member covariances. This is significantly different from standard transfer learning methods, as the focus is not on sharing information over tasks, but rather over system specifications. We demonstrate our approach on two benchmarks and a realistic wind farm setting. Our method significantly outperforms two baseline approaches, namely individual learning and joint learning where all samples are aggregated, in terms of the median and variance of the results.

Originele taal-2English
TitelProceedings of the 24th European Conference on Artificial Intelligence (ECAI 2020)
RedacteurenGiuseppe De Giacomo, Alejandro Catala, Bistra Dilkina, Michela Milano, Senen Barro, Alberto Bugarin, Jerome Lang
UitgeverijIOS Press
Pagina's1571-1578
Aantal pagina's8
Volume325
ISBN van elektronische versie9781643681009
ISBN van geprinte versie978-1-64368-100-9
DOI's
StatusPublished - 24 aug 2020
EvenementEuropean Conference on Artificial Intelligence (ECAI 2020) - Santiago De Compostela, Spain
Duur: 29 aug 20202 sep 2020
Congresnummer: 24
http://ecai2020.eu/

Publicatie series

NaamFrontiers in Artificial Intelligence and Applications

Conference

ConferenceEuropean Conference on Artificial Intelligence (ECAI 2020)
Verkorte titelECAI
Land/RegioSpain
StadSantiago De Compostela
Periode29/08/202/09/20
Internet adres

Vingerafdruk

Duik in de onderzoeksthema's van 'Fleet Control using Coregionalized Gaussian Process Policy Iteration'. Samen vormen ze een unieke vingerafdruk.

Citeer dit