Fed-EC: Bandwidth-Efficient Clustering-Based Federated Learning for Autonomous Visual Robot Navigation

Centralized learning requires data to be aggregated at a central server, which poses significant challenges in terms of data privacy and bandwidth consumption. Federated learning presents a compelling alternative, however, vanilla federated learning methods deployed in robotics aim to learn a single...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	IEEE robotics and automation letters Ročník 9; číslo 12; s. 11841 - 11848
Hlavní autori:	Gummadi, Shreya, Gasparino, Mateus V., Vasisht, Deepak, Chowdhary, Girish
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	IEEE 01.12.2024
Predmet:	Bandwidth Computational modeling Computer vision Data models Decentralized control Distributed robot systems Federated learning Navigation Predictive models Robot vision systems robotics in under-resourced settings Servers Training vision-based navigation
ISSN:	2377-3766, 2377-3766
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	Centralized learning requires data to be aggregated at a central server, which poses significant challenges in terms of data privacy and bandwidth consumption. Federated learning presents a compelling alternative, however, vanilla federated learning methods deployed in robotics aim to learn a single global model across robots that works ideally for all. But in practice one model may not be well suited for robots deployed in various environments. This letter proposes Federated-EmbedCluster (Fed-EC), a clustering-based federated learning framework that is deployed with vision based autonomous robot navigation in diverse outdoor environments. The framework addresses the key federated learning challenge of deteriorating model performance of a single global model due to the presence of non-IID data across real-world robots. Extensive real-world experiments validate that Fed-EC reduces the communication size by 23x for each robot while matching the performance of centralized learning for goal-oriented navigation and outperforms local learning. Fed-EC can transfer previously learnt models to new robots that join the cluster.
AbstractList	Centralized learning requires data to be aggregated at a central server, which poses significant challenges in terms of data privacy and bandwidth consumption. Federated learning presents a compelling alternative, however, vanilla federated learning methods deployed in robotics aim to learn a single global model across robots that works ideally for all. But in practice one model may not be well suited for robots deployed in various environments. This letter proposes Federated-EmbedCluster (Fed-EC), a clustering-based federated learning framework that is deployed with vision based autonomous robot navigation in diverse outdoor environments. The framework addresses the key federated learning challenge of deteriorating model performance of a single global model due to the presence of non-IID data across real-world robots. Extensive real-world experiments validate that Fed-EC reduces the communication size by 23x for each robot while matching the performance of centralized learning for goal-oriented navigation and outperforms local learning. Fed-EC can transfer previously learnt models to new robots that join the cluster.
Author	Chowdhary, Girish Gasparino, Mateus V. Gummadi, Shreya Vasisht, Deepak
Author_xml	– sequence: 1 givenname: Shreya orcidid: 0009-0002-1640-7617 surname: Gummadi fullname: Gummadi, Shreya email: gummadi4@illinois.edu organization: Field Robotics Engineering and Science Hub (FRESH), Illinois Autonomous Farm, University of Illinois at Urbana-Champaign, Urbana, IL, USA – sequence: 2 givenname: Mateus V. orcidid: 0000-0001-7900-8061 surname: Gasparino fullname: Gasparino, Mateus V. organization: Field Robotics Engineering and Science Hub (FRESH), Illinois Autonomous Farm, University of Illinois at Urbana-Champaign, Urbana, IL, USA – sequence: 3 givenname: Deepak orcidid: 0000-0003-3826-0978 surname: Vasisht fullname: Vasisht, Deepak organization: Siebel School of Computing and Data Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA – sequence: 4 givenname: Girish orcidid: 0000-0002-4657-307X surname: Chowdhary fullname: Chowdhary, Girish organization: Field Robotics Engineering and Science Hub (FRESH), Illinois Autonomous Farm, University of Illinois at Urbana-Champaign, Urbana, IL, USA
BookMark	eNp9kD1PwzAQhi1UJErpzsDgP5DijyRO2NqoBaQIpKpijS6JXYxSG9kOiH_flHaoGJjule6eu9NzjUbGGonQLSUzSkl-X67nM0ZYPONxngmRXaAx40JEXKTp6Cxfoan3H4QQmjDB82SM1Eq20bJ4wAsw7bduw3u0VEo3WpqAi673QTptttECvGzxMCwdhCGVEpwZGlhZh-d9sMbubO_xm_Y9dHhtaxvwC3zpLQRtzQ26VNB5OT3VCdqslpviKSpfH5-LeRk1LI1DRFPRJkS0rGaizrMkifMma6gSnHGSEaiB8gZInMiMKkJTJkkKKqYyVgBC8Qkix7WNs947qapPp3fgfipKqoOpajBVHUxVJ1MDkv5BGh1-fw4OdPcfeHcEtZTy7I5IOGEZ3wPINngr
CODEN	IRALC6
CitedBy_id	crossref_primary_10_1109_LRA_2025_3608659 crossref_primary_10_1109_LRA_2025_3573630
Cites_doi	10.1109/TCDS.2023.3239815 10.1109/TITS.2021.3081560 10.1109/IROS.2018.8594204 10.1109/ICRA48506.2021.9561936 10.1109/LRA.2019.2931179 10.1109/URAI.2018.8441797 10.1109/ACCESS.2020.2987642 10.1109/IV51971.2022.9827020 10.1109/ICRA57147.2024.10610436 10.1109/TNNLS.2019.2944481 10.1109/LRA.2020.2976321 10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00042 10.1109/TNNLS.2020.3015958 10.1109/GTSD.2018.8595590 10.15607/RSS.2022.XVIII.019 10.1109/MSP.2020.2975749 10.1109/ICRA48506.2021.9560791 10.15607/RSS.2021.XVII.019 10.1109/IJCNN48605.2020.9207469
ContentType	Journal Article
DBID	97E ESBDL RIA RIE AAYXX CITATION
DOI	10.1109/LRA.2024.3498778
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef
DatabaseTitle	CrossRef
DatabaseTitleList
Database_xml	– sequence: 1 dbid: RIE name: IEL url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	2377-3766
EndPage	11848
ExternalDocumentID	10_1109_LRA_2024_3498778 10753028
Genre	orig-research
GrantInformation_xml	– fundername: AIFARMS grantid: 1024178 – fundername: NSF-USDA COALESCE grantid: 2021-67021-34418
GroupedDBID	0R~ 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFS AGQYO AGSQL AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD ESBDL IFIPE IPLJI JAVBF KQ8 M43 M~E O9- OCL RIA RIE AAYXX CITATION
ID	FETCH-LOGICAL-c264t-167d507d2b27b985549c8c1f7323080aba13ca045e81f0162e06af41e4faa7f3
IEDL.DBID	RIE
ISICitedReferencesCount	1
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001409548200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2377-3766
IngestDate	Sat Nov 29 01:34:40 EST 2025 Tue Nov 18 22:30:59 EST 2025 Wed Aug 27 02:29:08 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	12
Language	English
License	https://creativecommons.org/licenses/by-nc-nd/4.0
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c264t-167d507d2b27b985549c8c1f7323080aba13ca045e81f0162e06af41e4faa7f3
ORCID	0000-0002-4657-307X 0009-0002-1640-7617 0000-0001-7900-8061 0000-0003-3826-0978
OpenAccessLink	https://ieeexplore.ieee.org/document/10753028
PageCount	8
ParticipantIDs	ieee_primary_10753028 crossref_citationtrail_10_1109_LRA_2024_3498778 crossref_primary_10_1109_LRA_2024_3498778
PublicationCentury	2000
PublicationDate	2024-12-01
PublicationDateYYYYMMDD	2024-12-01
PublicationDate_xml	– month: 12 year: 2024 text: 2024-12-01 day: 01
PublicationDecade	2020
PublicationTitle	IEEE robotics and automation letters
PublicationTitleAbbrev	LRA
PublicationYear	2024
Publisher	IEEE
Publisher_xml	– name: IEEE
References	ref12 ref15 ref14 ref31 ref30 ref11 ref33 Zhu (ref8) 2016 ref10 Bonawitz (ref20) 2019; 1 Li (ref25) 2020; 2 ref32 McMahan (ref13) 2017; 54 ref2 ref16 Hard (ref19) 2018 Zhao (ref26) 2018 Dennis (ref29) 2021 Kahn (ref4) 2021; 6 Konecn (ref18) 2016 Li (ref23) 2020 ref24 Ester (ref34) 1996 ref21 Chaplot (ref9) 2020 ref28 ref27 Shah (ref5) 2022 ref7 Cho (ref35) 2022 Gasparino (ref1) 2022; 7 ref3 ref6 Ghosh (ref17) 2020; 33 Mohri (ref22) 2019
References_xml	– ident: ref15 doi: 10.1109/TCDS.2023.3239815 – ident: ref31 doi: 10.1109/TITS.2021.3081560 – volume: 54 start-page: 1273 volume-title: Proc. 20th Int. Conf. Artif. Intell. Statist. year: 2017 ident: ref13 article-title: Communication-efficient learning of deep networks from decentralized data – ident: ref3 doi: 10.1109/IROS.2018.8594204 – ident: ref12 doi: 10.1109/ICRA48506.2021.9561936 – volume: 7 start-page: 10651 issue: 4 volume-title: IEEE Robot. Automat. Lett. year: 2022 ident: ref1 article-title: WayFAST: Navigation with predictive traversability in the field – start-page: 2611 volume-title: Proc. Int. Conf. Mach. Learn. year: 2021 ident: ref29 article-title: Heterogeneity for the win: One-shot federated clustering – ident: ref30 doi: 10.1109/LRA.2019.2931179 – ident: ref7 doi: 10.1109/URAI.2018.8441797 – ident: ref11 doi: 10.1109/ACCESS.2020.2987642 – year: 2018 ident: ref19 article-title: Federated learning for mobile keyboard prediction – ident: ref32 doi: 10.1109/IV51971.2022.9827020 – ident: ref2 doi: 10.1109/ICRA57147.2024.10610436 – year: 2022 ident: ref35 article-title: To federate or not to federate: Incentivizing client participation in federated learning – ident: ref24 doi: 10.1109/TNNLS.2019.2944481 – year: 2018 ident: ref26 article-title: Federated learning with non-IID data – volume-title: Proc. 8th Int. Conf. Learn. Representations year: 2020 ident: ref23 article-title: On the convergence of fedAvg on non-IID data – ident: ref14 doi: 10.1109/LRA.2020.2976321 – ident: ref28 doi: 10.1109/ISPA-BDCloud-SocialCom-SustainCom52081.2021.00042 – start-page: 4615 volume-title: Proc. Int. Conf. Mach. Learn. year: 2019 ident: ref22 article-title: Agnostic federated learning – start-page: 3357 volume-title: Proc. 2017 IEEE Int. Conf. Robot. Automat. year: 2016 ident: ref8 article-title: Target-driven visual navigation in indoor scenes using deep reinforcement learning – start-page: 12872 volume-title: Proc. 2020 IEEE/CVF Conf. Comput. Vis. Pattern Recognit. year: 2020 ident: ref9 article-title: Neural topological slam for visual navigation – ident: ref16 doi: 10.1109/TNNLS.2020.3015958 – ident: ref6 doi: 10.1109/GTSD.2018.8595590 – volume-title: Proc. Robot.: Sci. Syst. year: 2022 ident: ref5 article-title: ViKiNG: Vision-based kilometer-scale navigation with geographic hints doi: 10.15607/RSS.2022.XVIII.019 – volume: 33 start-page: 19586 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2020 ident: ref17 article-title: An efficient framework for clustered federated learning – volume: 1 start-page: 374 volume-title: Proc. Mach. Learn. Syst. year: 2019 ident: ref20 article-title: Towards federated learning at scale: System design – ident: ref21 doi: 10.1109/MSP.2020.2975749 – ident: ref33 doi: 10.1109/ICRA48506.2021.9560791 – start-page: 291 volume-title: Proc. 2nd Int. Conf. Knowl. Discov. Data Mining year: 1996 ident: ref34 article-title: A density-based algorithm for discovering clusters in large spatial databases with noise – ident: ref10 doi: 10.15607/RSS.2021.XVII.019 – volume: 6 start-page: 1312 issue: 2 volume-title: IEEE Robot. Automat. Lett. year: 2021 ident: ref4 article-title: BADGR: An autonomous self-supervised learning-based navigation system – volume: 2 start-page: 429 volume-title: Proc. Mach. Learn. Syst. year: 2020 ident: ref25 article-title: Federated optimization in heterogeneous networks – ident: ref27 doi: 10.1109/IJCNN48605.2020.9207469 – year: 2016 ident: ref18 article-title: Federated optimization: Distributed machine learning for on-device intelligence
SSID	ssj0001527395
Score	2.30905
Snippet	Centralized learning requires data to be aggregated at a central server, which poses significant challenges in terms of data privacy and bandwidth consumption....
SourceID	crossref ieee
SourceType	Enrichment Source Index Database Publisher
StartPage	11841
SubjectTerms	Bandwidth Computational modeling Computer vision Data models Decentralized control Distributed robot systems Federated learning Navigation Predictive models Robot vision systems robotics in under-resourced settings Servers Training vision-based navigation
Title	Fed-EC: Bandwidth-Efficient Clustering-Based Federated Learning for Autonomous Visual Robot Navigation
URI	https://ieeexplore.ieee.org/document/10753028
Volume	9
WOSCitedRecordID	wos001409548200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEL customDbUrl: eissn: 2377-3766 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001527395 issn: 2377-3766 databaseCode: RIE dateStart: 20160101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2377-3766 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0001527395 issn: 2377-3766 databaseCode: M~E dateStart: 20160101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwELZoxQADb0R5VB5YGNzm4TzM1laJGEqFqqrqFjl-QKUqQW1SNn47tpNCFpDYouicRPkS3Xfnu-8AuMciJDqFjzAnHsIktVHqYw_5IiVG_iQ17WPzcTCZhIsFeamb1U0vjBDCFJ-Jnj40e_k8Z6VOlak_PNBDbsIWaAWBXzVr_SRUtJQY8XZbkRbpj6cDFQA6uOdiFVnrQWoN19OYpWJcSXz8z4c4AUc1Z4SDCuRTsCeyM3DYUBI8BzIWHEWjRzikGf9Y8uINRUYcQl0LjlallkNQhmionBaHsVaQUCSTw1pe9RUq7goHZaFbHPJyA-fLTaluOc3TvIATujVCHHl2AWZxNBs9oXqEAmKK6RTI9gOuGB93UidIiS5JIyxktgxcFXqEFk2p7TKqaJ0IbanYnyMsn0psCywpDaR7CdpZnokrAKXjckwoc4XkmFNGPMKJLbHPhOdKwTugv3u5CavlxfWUi1ViwgyLJAqORMOR1HB0wMP3ivdKWuMP2wuNRMOuAuH6l_M34EAvr-pObkG7WJfiDuyzbbHcrLug9fwZdc3n8wVs68MD
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LS8NAEF60CurBZ8X63IMXD9vmsXmst7Y0VKxFSim9hc0-tFASaZP6993dpNqLgrcQJpuQL2G-mZ35BoB7LEKiU_gIc-IhTBIbJT72kC8SYuRPEtM-NhkEw2E4nZLXqlnd9MIIIUzxmWjqQ7OXzzNW6FSZ-sMDPeQm3AY7HsaOVbZr_aRUtJgY8dabkRZpDUZtFQI6uOliFVvrUWobzmdjmopxJtHRPx_jGBxWrBG2S5hPwJZIT8HBhpbgGZCR4KjXfYQdmvLPGc_fUc_IQ6i1YHdeaEEEZYg6ym1xGGkNCUUzOawEVt-gYq-wXeS6ySErlnAyWxbqlqMsyXI4pCsjxZGldTCOeuNuH1VDFBBTXCdHth9wxfm4kzhBQnRRGmEhs2XgquAjtGhCbZdRRexEaEvF_xxh-VRiW2BJaSDdc1BLs1RcACgdl2NCmSskx5wy4hFObIl9JjxXCt4ArfXLjVklMK7nXMxjE2hYJFZwxBqOuIKjAR6-r_goxTX-sK1rJDbsShAufzl_B_b645dBPHgaPl-Bfb1UWYVyDWr5ohA3YJet8tlycWs-oi8brMUZ
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Fed-EC%3A+Bandwidth-Efficient+Clustering-Based+Federated+Learning+for+Autonomous+Visual+Robot+Navigation&rft.jtitle=IEEE+robotics+and+automation+letters&rft.au=Gummadi%2C+Shreya&rft.au=Gasparino%2C+Mateus+V.&rft.au=Vasisht%2C+Deepak&rft.au=Chowdhary%2C+Girish&rft.date=2024-12-01&rft.pub=IEEE&rft.eissn=2377-3766&rft.volume=9&rft.issue=12&rft.spage=11841&rft.epage=11848&rft_id=info:doi/10.1109%2FLRA.2024.3498778&rft.externalDocID=10753028
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2377-3766&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2377-3766&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2377-3766&client=summon