Adaptive Asynchronous Control Using Meta-Learned Neural Ordinary Differential Equations
Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the applicability of those methods. In particular, we note...
Gespeichert in:
| Veröffentlicht in: | IEEE transactions on robotics Jg. 40; S. 403 - 420 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
New York
IEEE
01.01.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
| Schlagworte: | |
| ISSN: | 1552-3098, 1941-0468 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the applicability of those methods. In particular, we note two problems that jointly happen in many industrial systems: first, irregular/asynchronous observations and actions and, second, dramatic changes in environment dynamics from an episode to another (e.g .<inline-formula><tex-math notation="LaTeX">,</tex-math></inline-formula> varying payload inertial properties). We propose a general framework that overcomes those difficulties by meta-learning adaptive dynamics models for continuous-time prediction and control. The proposed approach is task-agnostic and can be adapted to new tasks in a straight-forward manner. We present evaluations in two different robot simulations and on a real industrial robot. |
|---|---|
| AbstractList | Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the applicability of those methods. In particular, we note two problems that jointly happen in many industrial systems: first, irregular/asynchronous observations and actions and, second, dramatic changes in environment dynamics from an episode to another (e.g .<inline-formula><tex-math notation="LaTeX">,</tex-math></inline-formula> varying payload inertial properties). We propose a general framework that overcomes those difficulties by meta-learning adaptive dynamics models for continuous-time prediction and control. The proposed approach is task-agnostic and can be adapted to new tasks in a straight-forward manner. We present evaluations in two different robot simulations and on a real industrial robot. Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the applicability of those methods. In particular, we note two problems that jointly happen in many industrial systems: first, irregular/asynchronous observations and actions and, second, dramatic changes in environment dynamics from an episode to another (e.g .[Formula Omitted] varying payload inertial properties). We propose a general framework that overcomes those difficulties by meta-learning adaptive dynamics models for continuous-time prediction and control. The proposed approach is task-agnostic and can be adapted to new tasks in a straight-forward manner. We present evaluations in two different robot simulations and on a real industrial robot. Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics settings. However, real-world robotics systems often present challenges that limit the applicability of those methods. In particular, we note two problems that jointly happen in many industrial systems: first, irregular/asynchronous observations and actions and, second, dramatic changes in environment dynamics from an episode to another (e.g ., varying payload inertial properties). We propose a general framework that overcomes those difficulties by meta-learning adaptive dynamics models for continuous-time prediction and control. The proposed approach is task-agnostic and can be adapted to new tasks in a straight-forward manner. We present evaluations in two different robot simulations and on a real industrial robot. |
| Author | Salehi, Achkan Doncieux, Stephane Ruhl, Steffen |
| Author_xml | – sequence: 1 givenname: Achkan orcidid: 0000-0002-7180-0123 surname: Salehi fullname: Salehi, Achkan email: achkan.salehi@sorbonne-universite.fr organization: Sorbonne Université, CNRS, ISIR, Paris, France – sequence: 2 givenname: Steffen orcidid: 0009-0002-6913-6822 surname: Ruhl fullname: Ruhl, Steffen email: ruehl@magazino.eu organization: Perception Team, Magazino GmbH, München, Germany – sequence: 3 givenname: Stephane orcidid: 0000-0003-1541-054X surname: Doncieux fullname: Doncieux, Stephane email: stephane.doncieux@sorbonne-universite.fr organization: Sorbonne Université, CNRS, ISIR, Paris, France |
| BackLink | https://hal.science/hal-04723191$$DView record in HAL |
| BookMark | eNpNkM9LMzEQhoMo-PPuwcOCJw_bb5JsdjfHUqt-0M-CKB7DdHeikZrUZFfwvzelIt9phpdnhpfnmO374Imxcw4TzkH_eXxYTgQIOZFS1FLBHjviuuIlVHW7n3elRClBt4fsOKU3AFFpkEfsedrjZnCfVEzTl-9eY_BhTMUs-CGGdfGUnH8p_tGA5YIweuqLexojrotl7J3H-FVcO2spkh9cTucfIw4u-HTKDiyuE539zBP2dDN_nN2Vi-Xt39l0UXai0UNZN43UKwTeEOoOGuyoQqtWK4sKFFLXcW2VbVH2q74GbBVVsgYOVve20yRP2NXu7yuuzSa691zJBHTmbrow2wyqRkiu-SfP7OWO3cTwMVIazFsYo8_1jNDZE4DkMlOwo7oYUopkf99yMFvVJqs2W9XmR3U-udidOCL6Dxdtq4SS32-yfJg |
| CODEN | ITREAE |
| Cites_doi | 10.15607/RSS.2021.XVII.011 10.1016/j.automatica.2004.11.034 10.5220/0009821603140323 10.1109/LRA.2022.3148438 10.1109/ICRA48506.2021.9560914 10.1016/j.automatica.2020.109421 10.1109/TCST.2021.3130225 10.4249/scholarpedia.4650 10.1109/LRA.2021.3057046 10.1109/tpami.2021.3079209 10.1016/j.robot.2020.103710 10.1109/LRA.2022.3144787 10.15607/RSS.2021.XVII.056 10.1109/LRA.2022.3192609 10.1177/0278364917710318 10.1201/9781439803905 10.1007/springerreference_179268 10.1109/IROS.2004.1389727 10.1109/SSCI47803.2020.9308468 10.1109/tnnls.2022.3175757 10.1177/02783649231165085 10.1561/2200000086 10.1016/0771-050X(80)90013-3 10.1109/ICRA46639.2022.9811559 10.1109/ICRA.2018.8463189 10.1023/A:1010091220143 |
| ContentType | Journal Article |
| Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 Copyright |
| Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024 – notice: Copyright |
| DBID | 97E RIA RIE AAYXX CITATION 7SC 7SP 7TB 8FD FR3 JQ2 L7M L~C L~D 1XC VOOES |
| DOI | 10.1109/TRO.2023.3326350 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Mechanical & Transportation Engineering Abstracts Technology Research Database Engineering Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access) |
| DatabaseTitle | CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Mechanical & Transportation Engineering Abstracts Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Engineering Research Database Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional |
| DatabaseTitleList | Technology Research Database |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering Computer Science |
| EISSN | 1941-0468 |
| EndPage | 420 |
| ExternalDocumentID | oai:HAL:hal-04723191v1 10_1109_TRO_2023_3326350 10288525 |
| Genre | orig-research |
| GrantInformation_xml | – fundername: VeriDream grantid: 951992 – fundername: European Union's H2020-EU.1.2.2 Research and Innovation |
| GroupedDBID | .DC 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AIBXA AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD F5P HZ~ H~9 IFIPE IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIE RNS VJK AAYXX CITATION 7SC 7SP 7TB 8FD FR3 JQ2 L7M L~C L~D 1XC VOOES |
| ID | FETCH-LOGICAL-c279t-67739ba017ea9c07ace4af5bbfa505aecc19f5f8a3dbd60a85e436010f9dfc9e3 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 1 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001141871000033&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1552-3098 |
| IngestDate | Tue Oct 14 20:14:23 EDT 2025 Mon Jun 30 06:17:32 EDT 2025 Sat Nov 29 01:47:32 EST 2025 Wed Aug 27 01:53:40 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Industrial robots Model Learning for Control learning and adaptive systems Robust/Adaptive Control of Robotic Systems |
| 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 Copyright: http://hal.archives-ouvertes.fr/licences/copyright |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c279t-67739ba017ea9c07ace4af5bbfa505aecc19f5f8a3dbd60a85e436010f9dfc9e3 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 |
| ORCID | 0000-0002-7180-0123 0009-0002-6913-6822 0000-0003-1541-054X |
| OpenAccessLink | https://hal.science/hal-04723191 |
| PQID | 2904600313 |
| PQPubID | 27625 |
| PageCount | 18 |
| ParticipantIDs | hal_primary_oai_HAL_hal_04723191v1 ieee_primary_10288525 crossref_primary_10_1109_TRO_2023_3326350 proquest_journals_2904600313 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-01-01 |
| PublicationDateYYYYMMDD | 2024-01-01 |
| PublicationDate_xml | – month: 01 year: 2024 text: 2024-01-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | New York |
| PublicationPlace_xml | – name: New York |
| PublicationTitle | IEEE transactions on robotics |
| PublicationTitleAbbrev | TRO |
| PublicationYear | 2024 |
| 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 | Wang (ref32) 2020; 21 ref12 ref15 Xu (ref39) 2018 ref11 Coumans (ref19) 2021 ref17 (ref10) 2023 (ref2) 2019 Vinyals (ref40) 2016 ref51 Duan (ref37) 2016 Salimans (ref42) 2017 ref46 ref45 ref48 ref47 Jiahao (ref16) 2023 ref44 ref49 Finn (ref8) 2017 ref7 Poli (ref50) 2020 Nagabandi (ref9) 2019 ref3 ref6 Rubinstein (ref28) 1999; 1 Deisenroth (ref4) 2011 ref35 Ghavamzadeh (ref34) 2001; 18 ref31 ref30 McGuire (ref43) 2009 ref1 Kelly (ref52) 2020 Chen (ref18) 2018 Schaul (ref36) 2010; 5 ref24 ref23 ref26 ref20 ref22 ref21 Tallec (ref33) 2019 Pinneri (ref5) 2021 Hafner (ref29) 2019 Song (ref41) 2020 ref27 Yildiz (ref13) 2021 Chua (ref25) 2018 Oh (ref38) 2020 Du (ref14) 2020 |
| References_xml | – ident: ref3 doi: 10.15607/RSS.2021.XVII.011 – ident: ref31 doi: 10.1016/j.automatica.2004.11.034 – start-page: 1049 volume-title: Proc. Conf. Robot Learn. year: 2021 ident: ref5 article-title: Sample-efficient cross-entropy method for real-time planning – ident: ref46 doi: 10.5220/0009821603140323 – year: 2016 ident: ref37 article-title: RL $^{2}$: Fast reinforcement learning via slow reinforcement learning – ident: ref11 doi: 10.1109/LRA.2022.3148438 – ident: ref20 article-title: Gazebo turtlebot3 simulation – ident: ref47 doi: 10.1109/ICRA48506.2021.9560914 – volume: 18 start-page: 186 volume-title: Proc. Int. Conf. Mach. Learn. year: 2001 ident: ref34 article-title: Continuous-time hierarchical reinforcement learning – ident: ref30 doi: 10.1016/j.automatica.2020.109421 – ident: ref27 doi: 10.1109/TCST.2021.3130225 – start-page: 19805 volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2020 ident: ref14 article-title: Model-based reinforcement learning for semi-Markov decision processes with neural ODEs – year: 2023 ident: ref10 article-title: Human-timescale adaptation in an open-ended task space – volume: 5 issue: 6 year: 2010 ident: ref36 article-title: Metalearning publication-title: Scholarpedia doi: 10.4249/scholarpedia.4650 – ident: ref12 doi: 10.1109/LRA.2021.3057046 – volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2018 ident: ref18 article-title: Neural ordinary differential equations – start-page: 3630 volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2016 ident: ref40 article-title: Matching networks for one shot learning – start-page: 2555 volume-title: Proc. Int. Conf. Mach. Learn. year: 2019 ident: ref29 article-title: Learning latent dynamics for planning from pixels – year: 2019 ident: ref2 article-title: Solving Rubiks cube with a robot hand – start-page: 465 volume-title: Proc. 28th Int. Conf. Mach. Learn. year: 2011 ident: ref4 article-title: PILCO: A model-based and data-efficient approach to policy search – start-page: 6096 volume-title: Proc. Int. Conf. Mach. Learn. year: 2019 ident: ref33 article-title: Making deep Q-learning methods robust to time discretization – volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2018 ident: ref39 article-title: Meta-gradient reinforcement learning – start-page: 21105 volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2020 ident: ref50 article-title: Hypersolvers: Toward fast continuous-depth models – year: 2021 ident: ref19 article-title: Pybullet, a python module for physics simulation for games, robotics and machine learning – ident: ref35 doi: 10.1109/tpami.2021.3079209 – ident: ref21 article-title: Robotic operating system – ident: ref7 doi: 10.1016/j.robot.2020.103710 – start-page: 12009 volume-title: Proc. Int. Conf. Mach. Learn. year: 2021 ident: ref13 article-title: Continuous-time model-based reinforcement learning – ident: ref15 doi: 10.1109/LRA.2022.3144787 – start-page: 1060 volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2020 ident: ref38 article-title: Discovering reinforcement learning algorithms – ident: ref49 doi: 10.15607/RSS.2021.XVII.056 – year: 2017 ident: ref42 article-title: Evolution strategies as a scalable alternative to reinforcement learning – ident: ref26 doi: 10.1109/LRA.2022.3192609 – ident: ref1 doi: 10.1177/0278364917710318 – volume-title: Conveyors: Application, Selection, and Integration year: 2009 ident: ref43 doi: 10.1201/9781439803905 – ident: ref23 doi: 10.1007/springerreference_179268 – ident: ref45 doi: 10.1109/IROS.2004.1389727 – volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2018 ident: ref25 article-title: Deep reinforcement learning in a handful of trials using probabilistic dynamics models – volume: 21 start-page: 1 issue: 198 year: 2020 ident: ref32 article-title: Reinforcement learning in continuous time and space: A stochastic control approach publication-title: J. Mach. Learn. Res. – ident: ref48 doi: 10.1109/SSCI47803.2020.9308468 – ident: ref51 doi: 10.1109/tnnls.2022.3175757 – ident: ref17 doi: 10.1177/02783649231165085 – ident: ref22 doi: 10.1561/2200000086 – start-page: 4370 volume-title: Proc. Annu. Conf. Neural Inf. Process. Syst. year: 2020 ident: ref52 article-title: Learning differential equations that are easy to solve – year: 2020 ident: ref41 article-title: ES-MAML: Simple Hessian-free meta learning publication-title: Proc. Int. Conf. Learn. Representations – year: 2019 ident: ref9 article-title: Learning to adapt in dynamic, real-world environments through meta-reinforcement learning publication-title: Proc. Int. Conf. Learn. Representations – ident: ref44 doi: 10.1016/0771-050X(80)90013-3 – start-page: 1126 volume-title: Proc. Int. Conf. Mach. Learn. year: 2017 ident: ref8 article-title: Model-agnostic meta-learning for fast adaptation of deep networks – ident: ref6 doi: 10.1109/ICRA46639.2022.9811559 – ident: ref24 doi: 10.1109/ICRA.2018.8463189 – start-page: 2251 volume-title: Proc. Conf. Robot Learn. year: 2023 ident: ref16 article-title: Online dynamics learning for predictive control with an application to aerial robots – volume: 1 start-page: 127 issue: 2 year: 1999 ident: ref28 article-title: The cross-entropy method for combinatorial and continuous optimization publication-title: Methodol. Comput. Appl. Probability doi: 10.1023/A:1010091220143 |
| SSID | ssj0024903 |
| Score | 2.4360971 |
| Snippet | Model-based reinforcement learning and control have demonstrated great potential in various sequential decision making problem domains, including in robotics... |
| SourceID | hal proquest crossref ieee |
| SourceType | Open Access Repository Aggregation Database Index Database Publisher |
| StartPage | 403 |
| SubjectTerms | Adaptive control Adaptive systems Computer Science Differential equations Industrial robots learning and adaptive systems Machine Learning Metalearning model learning for control Reinforcement learning Robot control Robot learning Robotics robust/adaptive control of robotic systems Service robots |
| Title | Adaptive Asynchronous Control Using Meta-Learned Neural Ordinary Differential Equations |
| URI | https://ieeexplore.ieee.org/document/10288525 https://www.proquest.com/docview/2904600313 https://hal.science/hal-04723191 |
| Volume | 40 |
| WOSCitedRecordID | wos001141871000033&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 Electronic Library (IEL) customDbUrl: eissn: 1941-0468 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0024903 issn: 1552-3098 databaseCode: RIE dateStart: 20040101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ1LSwMxEIAHWzzowWfFapUgXjykTfeVzbHUlh60FanY25LNJihIW_sC_72Z7FYr4sHbsuyGkElmMsnMNwDXgbFGNlCSelIrGnAlaBqlHo1EGiomm1oUxSZ4vx-PRuKhSFZ3uTBaaxd8puv46O7ys4la4lFZA41hHHphCUqc8zxZ6xusJ1wZZESKUZ-JeH0nyURj-DioY5nwuu8je4X9sEGlF4yAdKVVfuljZ2S6-__s3gHsFbtJ0srFfwhbenwEuxuMwWN4bmVyijqNtOYfY4UsXOvsk3Yeok5cyAC51wtJHWpVZwR5HbbRgZ06mKtLbosaKlYXvJHOe84Gn1fgqdsZtnu0qKZAlcfFgkac-yKVdgVqKRTjUulAmjBNjbS7ICsp1RQmNLH0szSLmIxDHfjorhmRGSW0fwLl8WSsT4FEylg3T0YmQ9iLx-JUs9gYaRBWH_uqCjfr8U2mOTQjcc4GE4mVRYKySApZVOHKCuDrM6Rd91p3Cb5DkKXVEM1VswoVHO6NxvKRrkJtLbCkWHvzxBN42YtMyrM_fjuHHduFID9JqUF5MVvqC9hWq8XrfHbpptUnlCLLeA |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ1LSwMxEIAHq4J68C3WZxAvHlKz782xaKViHyIVewvZbIKC1NpWwX9vJrvVinjwtiy7IWSSmUwy8w3AaWiskQ2VpL7UioaJ4jSLM5_GPIsUk57mZbGJpNNJ-31-Wyaru1wYrbULPtM1fHR3-fmLesOjsnM0hmnkRxVYiMLQ94p0rW-0HneFkBEqRgPG0-mtJOPnvbtuDQuF14IA6SvshxWqPGIMpCuu8ksjOzNztfbPDq7DarmfJPViAmzAnB5swsoMZXALHuq5HKJWI_Xxx0AhDde6--SiCFInLmiAtPVEUgdb1TlBYodttGsnD2brksuyiorVBs-k8VrQwcfbcH_V6F00aVlPgSo_4RMaJ0nAM2nXoJZcsUQqHUoTZZmRdh9kZaU8biKTyiDP8pjJNNJhgA6b4blRXAc7MD94GehdILEy1tGTsckR9-KzNNMsNUYaxNWngarC2XR8xbDAZgjnbjAurCwEykKUsqjCiRXA12fIu27WWwLfIcrS6gjv3avCNg73TGPFSFfhYCowUa6-sfA5XvcilXLvj9-OYanZa7dE67pzsw_Ltjthca5yAPOT0Zs-hEX1Pnkaj47cFPsEUxbOvw |
| 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=Adaptive+Asynchronous+Control+Using+Meta-Learned+Neural+Ordinary+Differential+Equations&rft.jtitle=IEEE+transactions+on+robotics&rft.au=Salehi%2C+Achkan&rft.au=Ruhl%2C+Steffen&rft.au=Doncieux%2C+Stephane&rft.date=2024-01-01&rft.pub=IEEE&rft.issn=1552-3098&rft.volume=40&rft.spage=403&rft.epage=420&rft_id=info:doi/10.1109%2FTRO.2023.3326350&rft.externalDocID=10288525 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1552-3098&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1552-3098&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1552-3098&client=summon |