Towards Generalization in Target-Driven Visual Navigation by Using Deep Reinforcement Learning

Among the main challenges in robotics, target-driven visual navigation has gained increasing interest in recent years. In this task, an agent has to navigate in an environment to reach a user specified target, only through vision. Recent fruitful approaches rely on deep reinforcement learning, which...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE transactions on robotics Vol. 36; no. 5; pp. 1546 - 1561
Main Authors:	Devo, Alessandro, Mezzetti, Giacomo, Costante, Gabriele, Fravolini, Mario L., Valigi, Paolo
Format:	Journal Article
Language:	English
Published:	New York IEEE 01.10.2020 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:	Computer simulation Deep learning Deep learning in robotics and automation Machine learning Navigation Robotics Simultaneous localization and mapping target-driven visual navigation Task analysis Training visual learning visual-based navigation Visualization
ISSN:	1552-3098, 1941-0468
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Among the main challenges in robotics, target-driven visual navigation has gained increasing interest in recent years. In this task, an agent has to navigate in an environment to reach a user specified target, only through vision. Recent fruitful approaches rely on deep reinforcement learning, which has proven to be an effective framework to learn navigation policies. However, current state-of-the-art methods require to retrain, or at least fine-tune, the model for every new environment and object. In real scenarios, this operation can be extremely challenging or even dangerous. For these reasons, we address generalization in target-driven visual navigation by proposing a novel architecture composed of two networks, both exclusively trained in simulation. The first one has the objective of exploring the environment, while the other one of locating the target. They are specifically designed to work together, while separately trained to help generalization. In this article, we test our agent in both simulated and real scenarios, and validate its generalization capabilities through extensive experiments with previously unseen goals and unknown mazes, even much larger than the ones used for training.
AbstractList	Among the main challenges in robotics, target-driven visual navigation has gained increasing interest in recent years. In this task, an agent has to navigate in an environment to reach a user specified target, only through vision. Recent fruitful approaches rely on deep reinforcement learning, which has proven to be an effective framework to learn navigation policies. However, current state-of-the-art methods require to retrain, or at least fine-tune, the model for every new environment and object. In real scenarios, this operation can be extremely challenging or even dangerous. For these reasons, we address generalization in target-driven visual navigation by proposing a novel architecture composed of two networks, both exclusively trained in simulation. The first one has the objective of exploring the environment, while the other one of locating the target. They are specifically designed to work together, while separately trained to help generalization. In this article, we test our agent in both simulated and real scenarios, and validate its generalization capabilities through extensive experiments with previously unseen goals and unknown mazes, even much larger than the ones used for training.
Author	Devo, Alessandro Mezzetti, Giacomo Costante, Gabriele Fravolini, Mario L. Valigi, Paolo
Author_xml	– sequence: 1 givenname: Alessandro orcidid: 0000-0001-7522-6264 surname: Devo fullname: Devo, Alessandro email: alessandro.devo@studenti.unipg.it organization: Department of Engineering, University of Perugia, Perugia, Italy – sequence: 2 givenname: Giacomo surname: Mezzetti fullname: Mezzetti, Giacomo email: giacomomezzet@gmail.com organization: Department of Engineering, University of Perugia, Perugia, Italy – sequence: 3 givenname: Gabriele orcidid: 0000-0002-8417-9372 surname: Costante fullname: Costante, Gabriele email: gabriele.costante@unipg.it organization: Department of Engineering, University of Perugia, Perugia, Italy – sequence: 4 givenname: Mario L. surname: Fravolini fullname: Fravolini, Mario L. email: mario.fravolini@unipg.it organization: Department of Engineering, University of Perugia, Perugia, Italy – sequence: 5 givenname: Paolo orcidid: 0000-0002-0486-7678 surname: Valigi fullname: Valigi, Paolo email: paolo.valigi@unipg.it organization: Department of Engineering, University of Perugia, Perugia, Italy
BookMark	eNp9kM1LAzEQxYNUsFbvgpeA56352o8cpdUqFAtl69EluzspKW22JttK_etN3eLBg6cZeL83j3mXqGcbCwjdUDKklMj7fD4bMsLIkEkpCGFnqE-loBERSdYLexyziBOZXaBL71cBEJLwPnrPm0_lao8nYMGptflSrWksNhbnyi2hjcbO7MHiN-N3ao1f1d4sO6Q84IU3donHAFs8B2N14yrYgG3xFJSzQbtC51qtPVyf5gAtnh7z0XM0nU1eRg_TqOJStpHiZcKAiSyjwLKMyLRSVOoakjrVcalrEcuYx1BrTWQSyyrlQpcySSpaUgDBB-iuu7t1zccOfFusmp2zIbJgQmRCpDzjgSIdVbnGewe62DqzUe5QUFIcWyxCi8WxxeLUYrAkfyyVaX_-b50y6_-Mt53RAMBvjqRBSij_BnPSgbQ
CODEN	ITREAE
CitedBy_id	crossref_primary_10_1007_s13198_021_01570_5 crossref_primary_10_1109_LRA_2024_3412638 crossref_primary_10_1109_ACCESS_2024_3413864 crossref_primary_10_1038_s41598_022_07264_7 crossref_primary_10_1109_LRA_2024_3511437 crossref_primary_10_1109_LRA_2024_3438588 crossref_primary_10_1109_TIE_2022_3203761 crossref_primary_10_1007_s11063_023_11190_8 crossref_primary_10_1109_LRA_2022_3154019 crossref_primary_10_1109_TPAMI_2024_3511936 crossref_primary_10_1016_j_neubiorev_2025_106282 crossref_primary_10_1109_LRA_2022_3178788 crossref_primary_10_1109_ACCESS_2023_3323801 crossref_primary_10_1109_TIM_2022_3158384 crossref_primary_10_1016_j_autcon_2023_105253 crossref_primary_10_1109_TIV_2024_3419846 crossref_primary_10_1016_j_robot_2022_104081 crossref_primary_10_1109_ACCESS_2025_3556894 crossref_primary_10_1007_s00521_024_09758_z crossref_primary_10_1109_LRA_2024_3455898 crossref_primary_10_1016_j_eij_2025_100680 crossref_primary_10_1177_17298806241292893 crossref_primary_10_1109_LRA_2024_3385700 crossref_primary_10_3389_frobt_2023_1255696 crossref_primary_10_1109_TITS_2024_3376579 crossref_primary_10_3390_s25113394 crossref_primary_10_1109_LRA_2020_3036597 crossref_primary_10_1007_s10489_022_03317_6 crossref_primary_10_1007_s12555_024_0045_7 crossref_primary_10_1109_JIOT_2025_3532164 crossref_primary_10_1016_j_aei_2023_101889 crossref_primary_10_1109_TNNLS_2021_3057424 crossref_primary_10_1109_LRA_2022_3150866 crossref_primary_10_1109_ACCESS_2024_3440183 crossref_primary_10_3390_photonics11070666 crossref_primary_10_1016_j_neucom_2024_129215 crossref_primary_10_1007_s00521_023_08991_2 crossref_primary_10_3390_s22218101 crossref_primary_10_1109_TRO_2025_3554252 crossref_primary_10_1109_LRA_2021_3133591 crossref_primary_10_1007_s10846_022_01646_9 crossref_primary_10_1007_s11063_022_10796_8 crossref_primary_10_1109_TIE_2024_3454534 crossref_primary_10_1109_LRA_2022_3140795 crossref_primary_10_1177_17298806241288669 crossref_primary_10_1088_1742_6596_2593_1_012003 crossref_primary_10_1109_TIP_2024_3459800 crossref_primary_10_3390_e25071007 crossref_primary_10_1007_s10846_024_02167_3 crossref_primary_10_1109_TASE_2023_3278740 crossref_primary_10_1109_TRO_2022_3207619 crossref_primary_10_1016_j_neunet_2023_06_007 crossref_primary_10_1016_j_neucom_2025_130191 crossref_primary_10_1109_TIE_2023_3337547 crossref_primary_10_1109_LRA_2022_3178810 crossref_primary_10_1109_TRO_2022_3191745 crossref_primary_10_1016_j_robot_2021_103799 crossref_primary_10_1109_LRA_2025_3547647 crossref_primary_10_1108_RIA_10_2023_0146 crossref_primary_10_1109_TRO_2024_3400932 crossref_primary_10_1109_ACCESS_2024_3493755 crossref_primary_10_1007_s12065_023_00817_3 crossref_primary_10_1109_LRA_2021_3068106
Cites_doi	10.1109/LRA.2019.2930426 10.1023/A:1008824626321 10.1109/IROS.2018.8593720 10.1109/ICRA.2017.7989385 10.1109/ICRA.2017.7989381 10.1109/IROS.2017.8202133 10.1109/CVPR.2016.91 10.1109/IROS.2018.8594249 10.1109/ICRA.2018.8460528 10.1109/ROBOT.1995.525695 10.1007/s10462-012-9365-8 10.1038/nature16961 10.1162/neco.1997.9.8.1735 10.1109/ICCV.2017.60 10.1109/CVPR.2017.769 10.1007/s11263-015-0816-y 10.1109/70.88137 10.1145/203330.203343 10.1109/ICRA.2018.8460655 10.1023/A:1007634325138 10.1109/WACV.2018.00097 10.1109/TRO.2018.2878318 10.1109/21.44033 10.1109/CVPR.2016.90 10.1177/0278364913495721 10.1109/LRA.2020.2965857 10.1109/ICRA.2019.8793493 10.1177/0278364919887447 10.1109/TRO.2017.2705103 10.1109/ICRA.2018.8460934 10.1038/nature14236 10.1109/70.938381 10.1007/978-3-319-46466-4_15 10.1109/TRO.2016.2624754
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020
DBID	97E RIA RIE AAYXX CITATION 7SC 7SP 7TB 8FD FR3 JQ2 L7M L~C L~D
DOI	10.1109/TRO.2020.2994002
DatabaseName	IEEE All-Society Periodicals Package (ASPP) 2005–Present 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
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
EISSN	1941-0468
EndPage	1561
ExternalDocumentID	10_1109_TRO_2020_2994002 9102361
Genre	orig-research
GrantInformation_xml	– fundername: Università degli Studi di Perugia; University of Perugia grantid: RICBA17MRF; RICBA18MF funderid: 10.13039/501100010607
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
ID	FETCH-LOGICAL-c399t-a3b62e24881e288097ca19fde6d7f5bfd459535edff09659c734fb966c1b1ee43
IEDL.DBID	RIE
ISICitedReferencesCount	81
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000576260400011&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	Sun Jun 29 16:02:01 EDT 2025 Sat Nov 29 01:47:27 EST 2025 Tue Nov 18 20:53:11 EST 2025 Wed Aug 27 02:31:54 EDT 2025
IsDoiOpenAccess	false
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	5
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-c399t-a3b62e24881e288097ca19fde6d7f5bfd459535edff09659c734fb966c1b1ee43
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-8417-9372 0000-0002-0486-7678 0000-0001-7522-6264
OpenAccessLink	https://doi.org/10.5281/zenodo.4785793
PQID	2448447383
PQPubID	27625
PageCount	16
ParticipantIDs	proquest_journals_2448447383 crossref_primary_10_1109_TRO_2020_2994002 ieee_primary_9102361 crossref_citationtrail_10_1109_TRO_2020_2994002
PublicationCentury	2000
PublicationDate	2020-Oct. 2020-10-00 20201001
PublicationDateYYYYMMDD	2020-10-01
PublicationDate_xml	– month: 10 year: 2020 text: 2020-Oct.
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on robotics
PublicationTitleAbbrev	TRO
PublicationYear	2020
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	silver (ref28) 2016; 529 ref12 ref15 anderson (ref52) 2018 wang (ref48) 2016 ref11 ref10 mirowski (ref14) 2016 sutton (ref25) 2018 ref17 ref16 sadeghi (ref31) 2016 ref19 kipf (ref40) 2016 bruce (ref35) 2017 packer (ref24) 2018 ref51 ref50 greydanus (ref53) 0 ref46 ref45 farebrother (ref23) 2018 ref47 clemente (ref42) 2017 ref49 ref8 rupprecht (ref54) 2019 ref7 ref9 ref4 ref3 ref6 ref5 mnih (ref34) 0 ref37 ref36 mnih (ref29) 2015; 518 ref30 ref33 ref2 ref1 ref38 mnih (ref41) 2013 espeholt (ref44) 2018 babaeizadeh (ref43) 2016 kalashnikov (ref18) 0 ref26 ref20 ref22 ref21 ref27 levine (ref32) 2016; 17 jaderberg (ref13) 2016 yang (ref39) 2018
References_xml	– ident: ref36 doi: 10.1109/LRA.2019.2930426 – ident: ref3 doi: 10.1023/A:1008824626321 – year: 2017 ident: ref42 article-title: Efficient parallel methods for deep reinforcement learning – start-page: 1407 year: 2018 ident: ref44 article-title: Impala: Scalable distributed deep-rl with importance weighted actor-learner architectures publication-title: Int Conf Mach Learn – start-page: 1787 year: 0 ident: ref53 article-title: Visualizing and understanding Atari agents publication-title: Proc Int Conf Mach Learn – ident: ref22 doi: 10.1109/IROS.2018.8593720 – start-page: 1928 year: 0 ident: ref34 article-title: Asynchronous methods for deep reinforcement learning publication-title: Proc Int Conf Mach Learn – year: 2018 ident: ref52 article-title: On evaluation of embodied navigation agents – ident: ref17 doi: 10.1109/ICRA.2017.7989385 – year: 2019 ident: ref54 article-title: Finding and visualizing weaknesses of deep reinforcement learning agents – ident: ref21 doi: 10.1109/ICRA.2017.7989381 – year: 2018 ident: ref23 article-title: Generalization and regularization in DQN – year: 2016 ident: ref31 article-title: Cad2rl: Real single-image flight without a single real image – ident: ref49 doi: 10.1109/IROS.2017.8202133 – ident: ref9 doi: 10.1109/CVPR.2016.91 – ident: ref30 doi: 10.1109/IROS.2018.8594249 – year: 2018 ident: ref25 publication-title: Reinforcement Learning An Introduction – start-page: 651 year: 0 ident: ref18 article-title: Scalable deep reinforcement learning for vision-based robotic manipulation publication-title: Proc Conf Robot Learn – ident: ref50 doi: 10.1109/ICRA.2018.8460528 – ident: ref4 doi: 10.1109/ROBOT.1995.525695 – ident: ref7 doi: 10.1007/s10462-012-9365-8 – volume: 529 start-page: 484 year: 2016 ident: ref28 article-title: Mastering the game of go with deep neural networks and tree search publication-title: Nature doi: 10.1038/nature16961 – year: 2016 ident: ref14 article-title: Learning to navigate in complex environments – volume: 17 start-page: 1334 year: 2016 ident: ref32 article-title: End-to-end training of deep visuomotor policies publication-title: J Mach Learn Res – ident: ref46 doi: 10.1162/neco.1997.9.8.1735 – ident: ref38 doi: 10.1109/ICCV.2017.60 – ident: ref12 doi: 10.1109/CVPR.2017.769 – ident: ref45 doi: 10.1007/s11263-015-0816-y – ident: ref2 doi: 10.1109/70.88137 – year: 2018 ident: ref24 article-title: Assessing generalization in deep reinforcement learning – ident: ref26 doi: 10.1145/203330.203343 – year: 2016 ident: ref13 article-title: Reinforcement learning with unsupervised auxiliary tasks – ident: ref11 doi: 10.1109/ICRA.2018.8460655 – year: 2016 ident: ref40 article-title: Semi-supervised classification with graph convolutional networks – ident: ref27 doi: 10.1023/A:1007634325138 – ident: ref51 doi: 10.1109/WACV.2018.00097 – year: 2013 ident: ref41 article-title: Playing atari with deep reinforcement learning – ident: ref19 doi: 10.1109/TRO.2018.2878318 – year: 2018 ident: ref39 article-title: Visual semantic navigation using scene priors – ident: ref1 doi: 10.1109/21.44033 – ident: ref10 doi: 10.1109/CVPR.2016.90 – ident: ref16 doi: 10.1177/0278364913495721 – ident: ref15 doi: 10.1109/LRA.2020.2965857 – year: 2017 ident: ref35 article-title: One-shot reinforcement learning for robot navigation with interactive replay – ident: ref37 doi: 10.1109/ICRA.2019.8793493 – ident: ref20 doi: 10.1177/0278364919887447 – ident: ref5 doi: 10.1109/TRO.2017.2705103 – ident: ref33 doi: 10.1109/ICRA.2018.8460934 – volume: 518 start-page: 529 year: 2015 ident: ref29 article-title: Human-level control through deep reinforcement learning publication-title: Nature doi: 10.1038/nature14236 – ident: ref6 doi: 10.1109/70.938381 – ident: ref47 doi: 10.1007/978-3-319-46466-4_15 – ident: ref8 doi: 10.1109/TRO.2016.2624754 – year: 2016 ident: ref48 article-title: Sample efficient actor-critic with experience replay – year: 2016 ident: ref43 article-title: Ga3c: GPU-based a3c for deep reinforcement learning
SSID	ssj0024903
Score	2.610255
Snippet	Among the main challenges in robotics, target-driven visual navigation has gained increasing interest in recent years. In this task, an agent has to navigate...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1546
SubjectTerms	Computer simulation Deep learning Deep learning in robotics and automation Machine learning Navigation Robotics Simultaneous localization and mapping target-driven visual navigation Task analysis Training visual learning visual-based navigation Visualization
Title	Towards Generalization in Target-Driven Visual Navigation by Using Deep Reinforcement Learning
URI	https://ieeexplore.ieee.org/document/9102361 https://www.proquest.com/docview/2448447383
Volume	36
WOSCitedRecordID	wos000576260400011&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/eLvHCXMwlV3PS8MwFA5zeNCDv6Y4nZKDF8FsS5s2zVGc4kGmjCo7Wdr0RQbSjXUb-N-bpOkUFMFbDy-h5Mt7eS9573sIXQCnwCSNSMrCkLBIUJLyUBEFCnwlPYioLRR-4MNhNB6Lpwa6WtfCAIBNPoOu-bRv-flULs1VWU8YngET62xwHla1Wl-8esJ2QTaMYsTvi6h-kuyLXjx61IGg1-9q08vqC5T6CLI9VX4YYnu63O3-77_20I7zIvF1Bfs-akBxgLa_cQu20GtsE2JL7IilXb0lnhQ4ttnfZDA3lg6_TMqlnmuYrizbhhbJPrDNJMADgBkegSVXlfYeETs-1rdD9Hx3G9_cE9dMgUjtgyxI6mehB57WVwqeVlrBZUqFyiHMuQoylbNABH4AuVKGEEZI7jOV6WBI0owCMP8INYtpAccI51mUg0q18yUp0zvRMOinMgxkGkkdf2Vt1KvXN5GOadw0vHhPbMTRF4lGJDGIJA6RNrpcj5hVLBt_yLYMAms5t_ht1KkhTJwalon2XSLGuI7CT34fdYq2zNxVdl4HNRfzJZyhTblaTMr5ud1hnx7Zz5I
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB6kCurBt1itmoMXwbTNbvaRo6hFsVaRVXpy2c1OpCBt6UPw35uk2Soogrc9THaXTGYyXzLzDcAJRgy5ZDHNeBhSHgtGsyhUVKFCX0kPY2YLhdtRpxN3u-JhAc7mtTCIaJPPsG4e7V1-MZBTc1TWEIZnwGCdRdM5y1VrfTHrCdsH2XCKUb8p4vJSsikayeO9hoJes66dLy-PUMpNyHZV-eGK7f7SWv_fn23AmosjyflM8ZuwgP0tWP3GLrgNL4lNiR0TRy3tKi5Jr08Sm_9NL0fG15Hn3niq39XJ3i3fhhbJP4jNJSCXiEPyiJZeVdqTROIYWV934Kl1lVxcU9dOgUodhUxo5uehh562WIaeNlsRyYwJVWBYRCrIVcEDEfgBFkoZShghI5-rXMMhyXKGyP1dqPQHfdwDUuRxgSrT4ZdkXK9Fw6GfyTCQWSw1Asur0CjnN5WOa9y0vHhLLeZoilRrJDUaSZ1GqnA6HzGc8Wz8IbttNDCXc5NfhVqpwtQZ4jjV0UvMeaRx-P7vo45h-Tq5a6ftm87tAayY78xy9WpQmYymeAhL8n3SG4-O7Gr7BBg_0ts
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=Towards+Generalization+in+Target-Driven+Visual+Navigation+by+Using+Deep+Reinforcement+Learning&rft.jtitle=IEEE+transactions+on+robotics&rft.au=Devo%2C+Alessandro&rft.au=Mezzetti%2C+Giacomo&rft.au=Costante%2C+Gabriele&rft.au=Fravolini%2C+Mario+L.&rft.date=2020-10-01&rft.pub=IEEE&rft.issn=1552-3098&rft.volume=36&rft.issue=5&rft.spage=1546&rft.epage=1561&rft_id=info:doi/10.1109%2FTRO.2020.2994002&rft.externalDocID=9102361
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