Bridging the Model-Reality Gap With Lipschitz Network Adaptation

As robots venture into the real world, they are subject to unmodeled dynamics and disturbances. Traditional model-based control approaches have been proven successful in relatively static and known operating environments. However, when an accurate model of the robot is not available, model-based des...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE robotics and automation letters Ročník 7; číslo 1; s. 642 - 649
Hlavní autoři:	Zhou, Siqi, Pereida, Karime, Zhao, Wenda, Schoellig, Angela P.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Piscataway IEEE 01.01.2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Adaptation Adaptation models Aerodynamics Computational modeling Control systems design deep learning methods Dynamic stability Dynamical systems Hovering Lipschitz condition Machine learning for robot control Model reference adaptive control Neural networks Nonlinear control Nonlinear dynamical systems Nonlinear dynamics Nonlinear systems Pendulums Predictive models Robots robust/adaptive control Uncertainty
ISSN:	2377-3766, 2377-3766
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	As robots venture into the real world, they are subject to unmodeled dynamics and disturbances. Traditional model-based control approaches have been proven successful in relatively static and known operating environments. However, when an accurate model of the robot is not available, model-based design can lead to suboptimal and even unsafe behaviour. In this work, we propose a method that bridges the model-reality gap and enables the application of model-based approaches even if dynamic uncertainties are present. In particular, we present a learning-based model reference adaptation approach that makes a robot system, with possibly uncertain dynamics, behave as a predefined reference model. In turn, the reference model can be used for model-based controller design. In contrast to typical model reference adaptation control approaches, we leverage the representative power of neural networks to capture highly nonlinear dynamics uncertainties and guarantee stability by encoding a certifying Lipschitz condition in the architectural design of a special type of neural network called the Lipschitz network. Our approach applies to a general class of nonlinear control-affine systems even when our prior knowledge about the true robot system is limited. We demonstrate our approach in flying inverted pendulum experiments, where an off-the-shelf quadrotor is challenged to balance an inverted pendulum while hovering or tracking circular trajectories.
AbstractList	As robots venture into the real world, they are subject to unmodeled dynamics and disturbances. Traditional model-based control approaches have been proven successful in relatively static and known operating environments. However, when an accurate model of the robot is not available, model-based design can lead to suboptimal and even unsafe behaviour. In this work, we propose a method that bridges the model-reality gap and enables the application of model-based approaches even if dynamic uncertainties are present. In particular, we present a learning-based model reference adaptation approach that makes a robot system, with possibly uncertain dynamics, behave as a predefined reference model. In turn, the reference model can be used for model-based controller design. In contrast to typical model reference adaptation control approaches, we leverage the representative power of neural networks to capture highly nonlinear dynamics uncertainties and guarantee stability by encoding a certifying Lipschitz condition in the architectural design of a special type of neural network called the Lipschitz network. Our approach applies to a general class of nonlinear control-affine systems even when our prior knowledge about the true robot system is limited. We demonstrate our approach in flying inverted pendulum experiments, where an off-the-shelf quadrotor is challenged to balance an inverted pendulum while hovering or tracking circular trajectories.
Author	Pereida, Karime Zhao, Wenda Schoellig, Angela P. Zhou, Siqi
Author_xml	– sequence: 1 givenname: Siqi orcidid: 0000-0001-7240-546X surname: Zhou fullname: Zhou, Siqi email: siqi.zhou@robotics.utias.utoronto.ca organization: Dynamic Systems Lab (www.dynsyslab.org), Institute for Aerospace Studies, University of Toronto, Toronto, Canada – sequence: 2 givenname: Karime orcidid: 0000-0002-8576-2164 surname: Pereida fullname: Pereida, Karime email: karime.pereida@robotics.utias. utoronto.ca organization: Dynamic Systems Lab (www.dynsyslab.org), Institute for Aerospace Studies, University of Toronto, Toronto, Canada – sequence: 3 givenname: Wenda orcidid: 0000-0002-4288-0582 surname: Zhao fullname: Zhao, Wenda email: wenda.zhao@robotics.utias.utoronto.ca organization: Dynamic Systems Lab (www.dynsyslab.org), Institute for Aerospace Studies, University of Toronto, Toronto, Canada – sequence: 4 givenname: Angela P. orcidid: 0000-0003-4012-4668 surname: Schoellig fullname: Schoellig, Angela P. email: schoellig@utias.utoronto. ca organization: Dynamic Systems Lab (www.dynsyslab.org), Institute for Aerospace Studies, University of Toronto, Toronto, Canada
BookMark	eNp9kMFLwzAUh4NMcM7dBS8Fz50vSZM0N6foFKbCUDyGNE23zNrWNEPmX29Hh4gH3-W9w-977_Edo0FVVxahUwwTjEFezBfTCQGCJxRTzGV6gIaEChFTwfng13yExm27BgDMiKCSDdHllXf50lXLKKxs9FDntowXVpcubKOZbqJXF1bR3DWtWbnwFT3a8Fn7t2ia6ybo4OrqBB0WumzteN9H6OX25vn6Lp4_ze6vp_PYEIlDTFOa2SLRxgrDMkmSQkhGIe9Kc0lyIUiCDSSFSUBzjZm0Sa4zoFoQktOMjtB5v7fx9cfGtkGt642vupOKcEgZ4ymDLsX7lPF123pbKOP6P4PXrlQY1E6Y6oSpnTC1F9aB8AdsvHvXfvsfctYjzlr7E5eckgQY_QbNL3bC
CODEN	IRALC6
CitedBy_id	crossref_primary_10_3390_app122010277 crossref_primary_10_1109_TRO_2024_3381554 crossref_primary_10_1007_s10723_023_09663_0 crossref_primary_10_1109_LRA_2024_3368969 crossref_primary_10_1016_j_ifacol_2023_10_053
Cites_doi	10.1002/acs.2397 10.1016/0005-1098(94)00161-B 10.1049/ip-cta:19951613 10.1177/0278364920953902 10.1002/rob.21587 10.1109/CDC40024.2019.9029173 10.1109/CDC.2007.4434184 10.1109/LRA.2019.2901638 10.23919/ECC.2019.8796140 10.1121/1.399905 10.1109/ICRA40945.2020.9196744 10.1109/TIE.2016.2552151 10.1109/9.384214 10.1207/s15516709cog1603_1 10.1109/LRA.2018.2847405 10.1109/ICRA.2019.8794351 10.1109/ICRA.2011.5980244 10.1016/j.jprocont.2009.11.007 10.1137/1.9780898719376 10.1146/annurev-control-042920-020211 10.1109/ICRA.2018.8460528 10.1109/TNNLS.2014.2319052
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
DBID	97E RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
DOI	10.1109/LRA.2021.3131698
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Electronic Library Online CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE Xplore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	2377-3766
EndPage	649
ExternalDocumentID	10_1109_LRA_2021_3131698 9632405
Genre	orig-research
GrantInformation_xml	– fundername: CIFAR AI Chairs Program – fundername: Natural Sciences and Engineering Research Council of Canada funderid: 10.13039/501100000038 – fundername: Canada Research Chairs funderid: 10.13039/501100001804
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 IFIPE IPLJI JAVBF KQ8 M43 M~E O9- OCL RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c291t-383bef4ace7c5b924f79530dddda692d77241c04fc40a6a159e4dab03a722d3b3
IEDL.DBID	RIE
ISICitedReferencesCount	5
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000728922700005&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	Sun Jun 29 15:19:55 EDT 2025 Tue Nov 18 22:43:58 EST 2025 Sat Nov 29 06:03:13 EST 2025 Wed Aug 27 05:07:55 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	1
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c291t-383bef4ace7c5b924f79530dddda692d77241c04fc40a6a159e4dab03a722d3b3
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-8576-2164 0000-0003-4012-4668 0000-0001-7240-546X 0000-0002-4288-0582
PQID	2608556850
PQPubID	4437225
PageCount	8
ParticipantIDs	crossref_primary_10_1109_LRA_2021_3131698 ieee_primary_9632405 crossref_citationtrail_10_1109_LRA_2021_3131698 proquest_journals_2608556850
PublicationCentury	2000
PublicationDate	2022-Jan. 2022-1-00 20220101
PublicationDateYYYYMMDD	2022-01-01
PublicationDate_xml	– month: 01 year: 2022 text: 2022-Jan.
PublicationDecade	2020
PublicationPlace	Piscataway
PublicationPlace_xml	– name: Piscataway
PublicationTitle	IEEE robotics and automation letters
PublicationTitleAbbrev	LRA
PublicationYear	2022
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 ref12 ref14 Zhou (ref2) 1998 ref10 ref1 ref17 ref16 Hovakimyan (ref11) 2010 ref19 ref18 Fazlyab (ref20) 2019 ref24 ref26 ref25 ref22 ref21 ref27 ref8 ref7 ref9 ref4 ref3 Anil (ref6) 2019 ref5 Khalil (ref23) 2002 Xie (ref15) 2021
References_xml	– ident: ref4 doi: 10.1002/acs.2397 – ident: ref25 doi: 10.1016/0005-1098(94)00161-B – volume-title: Proc. Int. Conf. Mach. Learn. (ICML) year: 2021 ident: ref15 article-title: Deep reinforcement learning amidst lifelong non-stationarity – ident: ref18 doi: 10.1049/ip-cta:19951613 – ident: ref14 doi: 10.1177/0278364920953902 – volume-title: Essentials of Robust Control year: 1998 ident: ref2 – start-page: 11427 volume-title: Proc. Conf. Neural Inf. Process. Syst. year: 2019 ident: ref20 article-title: Efficient and accurate estimation of lipschitz constants for deep neural networks – ident: ref12 doi: 10.1002/rob.21587 – ident: ref19 doi: 10.1109/CDC40024.2019.9029173 – ident: ref24 doi: 10.1109/CDC.2007.4434184 – ident: ref13 doi: 10.1109/LRA.2019.2901638 – ident: ref22 doi: 10.23919/ECC.2019.8796140 – ident: ref3 doi: 10.1121/1.399905 – ident: ref27 doi: 10.1109/ICRA40945.2020.9196744 – ident: ref7 doi: 10.1109/TIE.2016.2552151 – ident: ref17 doi: 10.1109/9.384214 – ident: ref16 doi: 10.1207/s15516709cog1603_1 – ident: ref9 doi: 10.1109/LRA.2018.2847405 – start-page: 291 volume-title: Proc. Int. Conf. Mach. Learn. year: 2019 ident: ref6 article-title: Sorting out lipschitz function approximation – ident: ref21 doi: 10.1109/ICRA.2019.8794351 – volume-title: Nonlinear Systems Third Edition year: 2002 ident: ref23 – ident: ref26 doi: 10.1109/ICRA.2011.5980244 – ident: ref8 doi: 10.1016/j.jprocont.2009.11.007 – volume-title: Adaptive Control Theory: Guaranteed Robustness with Fast Adaptation year: 2010 ident: ref11 doi: 10.1137/1.9780898719376 – ident: ref1 doi: 10.1146/annurev-control-042920-020211 – ident: ref10 doi: 10.1109/ICRA.2018.8460528 – ident: ref5 doi: 10.1109/TNNLS.2014.2319052
SSID	ssj0001527395
Score	2.2591887
Snippet	As robots venture into the real world, they are subject to unmodeled dynamics and disturbances. Traditional model-based control approaches have been proven...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	642
SubjectTerms	Adaptation Adaptation models Aerodynamics Computational modeling Control systems design deep learning methods Dynamic stability Dynamical systems Hovering Lipschitz condition Machine learning for robot control Model reference adaptive control Neural networks Nonlinear control Nonlinear dynamical systems Nonlinear dynamics Nonlinear systems Pendulums Predictive models Robots robust/adaptive control Uncertainty
Title	Bridging the Model-Reality Gap With Lipschitz Network Adaptation
URI	https://ieeexplore.ieee.org/document/9632405 https://www.proquest.com/docview/2608556850
Volume	7
WOSCitedRecordID	wos000728922700005&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: IEEE Xplore 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/eLvHCXMwlV09T8MwED21FQMMfBVEoVQeWJAIdZw4jjcKamEoFapAdIsc2xGVqrbqBwMDvx3bSQsIhEQmD7YUPTu5e3e-dwBnqQoybSy7R3jKvNAYWE_QIPQixoSyJcwydoXCXdbrxYMBfyjBxboWRmvtLp_pSzt0uXw1kUsbKmtyqy1uBUvLjLG8VusznmKVxDhdZSIxb3b7LcP_iG9oaeBHPP5meVwrlR__X2dUOjv_e51d2C6cR9TKd3sPSnq8D1tfJAWrcHVtK7DMEBnPDtlOZyOvr52zjW7FFD0PFy-oO5zObf7gDfXyW-CopcQ0T8ofwFOn_Xhz5xVdEjxJuL_wDMVMdRYKqZmkqaFTGeM0wMo8IuJEGfc59CUOMxliEQnjvuhQiRQHghGigjQ4hMp4MtZHgChWWFGqFBE89ImODRkzWxbHNAq0n2Y1aK4QTGQhIW47WYwSRyUwTwzmicU8KTCvwfl6xTSXz_hjbtVivJ5XwFuD-mqTkuL7mieGhcVWO43i499XncAmsYUKLlhSh8pittSnsCFfF8P5rAHl-_d2wx2gD1mQwlg
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH7IFNSDv6Y4f-bgRbAuTZO2uTnFqViHjIm7lTRJcTC2sU0P_vUmaTcVRbCnHBIoX9K-972X9z2Ak0wFuTaW3SM8izxqDKwnWEC9MIqEsiXMMnaFwknUasXdLn9cgLN5LYzW2l0-0-d26HL5aihfbaiszq22uBUsXWSUEr-o1vqMqFgtMc5muUjM60m7YRgg8Q0xDfyQx99sj2um8uMP7MxKc_1_L7QBa6X7iBrFfm_Cgh5sweoXUcEqXFzaGiwzRMa3Q7bXWd9ra-duoxsxQs-96QtKeqOJzSC8o1ZxDxw1lBgVaflteGped65uvbJPgicJ96eeIZmZzqmQOpIsM4QqjzgLsDKPCDlRxoGmvsQ0lxSLUBgHRlMlMhyIiBAVZMEOVAbDgd4FxLDCijGliODUJzo2dMxsWhyzMNB-ltegPkMwlaWIuO1l0U8dmcA8NZinFvO0xLwGp_MVo0JA44-5VYvxfF4Jbw0OZpuUll_YJDU8LLbqaQzv_b7qGJZvOw9Jmty17vdhhdiyBRc6OYDKdPyqD2FJvk17k_GRO0YfBg_Ebg
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=Bridging+the+Model-Reality+Gap+With+Lipschitz+Network+Adaptation&rft.jtitle=IEEE+robotics+and+automation+letters&rft.au=Zhou%2C+Siqi&rft.au=Pereida%2C+Karime&rft.au=Zhao%2C+Wenda&rft.au=Schoellig%2C+Angela+P.&rft.date=2022-01-01&rft.issn=2377-3766&rft.eissn=2377-3766&rft.volume=7&rft.issue=1&rft.spage=642&rft.epage=649&rft_id=info:doi/10.1109%2FLRA.2021.3131698&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_LRA_2021_3131698
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