Graph Soft Actor-Critic Reinforcement Learning for Large-Scale Distributed Multirobot Coordination

Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL) context. In this work, we model the interactions among the robots as a graph and propose a novel off-policy actor-critic MARL algorithm to train dis...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transaction on neural networks and learning systems Vol. 36; no. 1; pp. 665 - 676
Main Authors: Hu, Yifan, Fu, Junjie, Wen, Guanghui
Format: Journal Article
Language:English
Published: United States IEEE 01.01.2025
Subjects:
ISSN:2162-237X, 2162-2388, 2162-2388
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL) context. In this work, we model the interactions among the robots as a graph and propose a novel off-policy actor-critic MARL algorithm to train distributed coordination policies on the graph by leveraging the ability of information extraction of graph neural networks (GNNs). First, a new type of Gaussian policy parameterized by the GNNs is designed for distributed decision-making in continuous action spaces. Second, a scalable centralized value function network is designed based on a novel GNN-based value function decomposition technique. Then, based on the designed actor and the critic networks, a GNN-based MARL algorithm named graph soft actor-critic (G-SAC) is proposed and utilized to train the distributed policies in an effective and centralized fashion. Finally, two custom multirobot coordination environments are built, under which the simulation results are performed to empirically demonstrate both the sample efficiency and the scalability of G-SAC as well as the strong zero-shot generalization ability of the trained policy in large-scale multirobot coordination problems.
AbstractList Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL) context. In this work, we model the interactions among the robots as a graph and propose a novel off-policy actor-critic MARL algorithm to train distributed coordination policies on the graph by leveraging the ability of information extraction of graph neural networks (GNNs). First, a new type of Gaussian policy parameterized by the GNNs is designed for distributed decision-making in continuous action spaces. Second, a scalable centralized value function network is designed based on a novel GNN-based value function decomposition technique. Then, based on the designed actor and the critic networks, a GNN-based MARL algorithm named graph soft actor-critic (G-SAC) is proposed and utilized to train the distributed policies in an effective and centralized fashion. Finally, two custom multirobot coordination environments are built, under which the simulation results are performed to empirically demonstrate both the sample efficiency and the scalability of G-SAC as well as the strong zero-shot generalization ability of the trained policy in large-scale multirobot coordination problems.
Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL) context. In this work, we model the interactions among the robots as a graph and propose a novel off-policy actor-critic MARL algorithm to train distributed coordination policies on the graph by leveraging the ability of information extraction of graph neural networks (GNNs). First, a new type of Gaussian policy parameterized by the GNNs is designed for distributed decision-making in continuous action spaces. Second, a scalable centralized value function network is designed based on a novel GNN-based value function decomposition technique. Then, based on the designed actor and the critic networks, a GNN-based MARL algorithm named graph soft actor-critic (G-SAC) is proposed and utilized to train the distributed policies in an effective and centralized fashion. Finally, two custom multirobot coordination environments are built, under which the simulation results are performed to empirically demonstrate both the sample efficiency and the scalability of G-SAC as well as the strong zero-shot generalization ability of the trained policy in large-scale multirobot coordination problems.Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL) context. In this work, we model the interactions among the robots as a graph and propose a novel off-policy actor-critic MARL algorithm to train distributed coordination policies on the graph by leveraging the ability of information extraction of graph neural networks (GNNs). First, a new type of Gaussian policy parameterized by the GNNs is designed for distributed decision-making in continuous action spaces. Second, a scalable centralized value function network is designed based on a novel GNN-based value function decomposition technique. Then, based on the designed actor and the critic networks, a GNN-based MARL algorithm named graph soft actor-critic (G-SAC) is proposed and utilized to train the distributed policies in an effective and centralized fashion. Finally, two custom multirobot coordination environments are built, under which the simulation results are performed to empirically demonstrate both the sample efficiency and the scalability of G-SAC as well as the strong zero-shot generalization ability of the trained policy in large-scale multirobot coordination problems.
Author Wen, Guanghui
Fu, Junjie
Hu, Yifan
Author_xml – sequence: 1
  givenname: Yifan
  orcidid: 0000-0003-0756-6157
  surname: Hu
  fullname: Hu, Yifan
  email: yfhu@seu.edu.cn
  organization: School of Mathematics, Southeast University, Nanjing, China
– sequence: 2
  givenname: Junjie
  orcidid: 0000-0002-1528-8727
  surname: Fu
  fullname: Fu, Junjie
  email: fujunjie@seu.edu.cn
  organization: School of Mathematics, Southeast University, Nanjing, China
– sequence: 3
  givenname: Guanghui
  orcidid: 0000-0003-0070-8597
  surname: Wen
  fullname: Wen, Guanghui
  email: ghwen@seu.edu.cn
  organization: School of Mathematics, Southeast University, Nanjing, China
BackLink https://www.ncbi.nlm.nih.gov/pubmed/37948149$$D View this record in MEDLINE/PubMed
BookMark eNp9kM1O3DAURi1EBZTyAqiqvGSTwb-Js0RTSpEClTqzYBfZzjV1lbGntrPo2zcwA6pY4I0_Wefz1T0f0WGIARA6p2RBKWkv1_f33WrBCOMLzlkrOTlAJ4zWrGJcqcPX3Dwco7Ocf5P51ETWoj1Cx7xphaKiPUHmJuntL7yKruArW2KqlskXb_FP8MHFZGEDoeAOdAo-POL5CXc6PUK1snoE_NXnkryZCgz4bhqLT9HEgpcxpsEHXXwMn9AHp8cMZ_v7FK2_Xa-X36vux83t8qqrLGeiVC1zZmiMprWVzlFirDVGizm5mmkrpALDtZRsoFwSrUCTGQWAxjWi5fwUXey-3ab4Z4Jc-o3PFsZRB4hT7plSLRNcKDmjX_boZDYw9NvkNzr97V-0zADbATbFnBO4V4SS_kl__6y_f9Lf7_XPJfWmZH15NlCS9uP71c-7qp_3-W8Wp0Jyxf8BuPWUHw
CODEN ITNNAL
CitedBy_id crossref_primary_10_1109_TAI_2024_3461630
crossref_primary_10_3390_electronics14040820
crossref_primary_10_1109_TRO_2025_3582836
crossref_primary_10_3390_electronics14081686
crossref_primary_10_3389_frobt_2025_1492526
Cites_doi 10.1016/j.automatica.2016.06.024
10.1109/ICRA46639.2022.9812263
10.1109/IROS51168.2021.9635898
10.1109/IROS.2018.8593871
10.1109/MSP.2020.3016143
10.1109/TII.2012.2219061
10.1109/TAC.2004.834113
10.1109/tnnls.2022.3172168
10.48550/ARXIV.1706.03762
10.1609/aaai.v32i1.11794
10.1109/tnnls.2021.3104987
10.1109/LRA.2022.3146912
10.1109/ICRA46639.2022.9811744
10.1109/TNNLS.2020.2984944
10.1109/IROS51168.2021.9635836
10.1109/OJCOMS.2021.3081996
10.1109/TNNLS.2021.3095431
10.1007/978-3-030-60990-0_12
10.1109/TAC.2015.2465071
10.1109/LRA.2021.3074885
10.1109/TCYB.2020.3000264
10.48550/arXiv.1812.05905
10.1109/ICRA.2018.8461113
10.1109/TSMC.2019.2961753
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7X8
DOI 10.1109/TNNLS.2023.3329530
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList
PubMed
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISSN 2162-2388
EndPage 676
ExternalDocumentID 37948149
10_1109_TNNLS_2023_3329530
10314538
Genre orig-research
Journal Article
GrantInformation_xml – fundername: National Key Research and Development Program of China
  grantid: 2022YFA1004702
– fundername: General Joint Fund of the Equipment Advance Research Program of Ministry of Education
  grantid: 8091B022114
– fundername: National Natural Science Foundation of China
  grantid: 62173085; 62325304; U22B2046; 62088101; 62073079
  funderid: 10.13039/501100001809
GroupedDBID 0R~
4.4
5VS
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACIWK
ACPRK
AENEX
AFRAH
AGQYO
AGSQL
AHBIQ
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
EBS
EJD
IFIPE
IPLJI
JAVBF
M43
MS~
O9-
OCL
PQQKQ
RIA
RIE
RNS
AAYXX
CITATION
NPM
7X8
ID FETCH-LOGICAL-c324t-92fbd7ba16c5ff10bccbba4f10f62ac458eb3a552d1350a8ea016ceee7f74933
IEDL.DBID RIE
ISICitedReferencesCount 19
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001106580100001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2162-237X
2162-2388
IngestDate Sat Sep 27 18:48:52 EDT 2025
Wed Dec 10 14:04:40 EST 2025
Sat Nov 29 01:40:28 EST 2025
Tue Nov 18 22:30:35 EST 2025
Wed Aug 27 01:57:59 EDT 2025
IsPeerReviewed false
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-c324t-92fbd7ba16c5ff10bccbba4f10f62ac458eb3a552d1350a8ea016ceee7f74933
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0003-0756-6157
0000-0002-1528-8727
0000-0003-0070-8597
PMID 37948149
PQID 2889243485
PQPubID 23479
PageCount 12
ParticipantIDs proquest_miscellaneous_2889243485
pubmed_primary_37948149
ieee_primary_10314538
crossref_primary_10_1109_TNNLS_2023_3329530
crossref_citationtrail_10_1109_TNNLS_2023_3329530
PublicationCentury 2000
PublicationDate 2025-Jan.
2025-1-00
2025-Jan
20250101
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – month: 01
  year: 2025
  text: 2025-Jan.
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle IEEE transaction on neural networks and learning systems
PublicationTitleAbbrev TNNLS
PublicationTitleAlternate IEEE Trans Neural Netw Learn Syst
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
References ref35
ref12
ref15
ref14
ref11
Lowe (ref28)
ref10
Zhu (ref18) 2022
ref2
ref1
ref16
ref19
Schulman (ref13) 2017
Green (ref46) 2019
Wang (ref26)
Nayak (ref17)
ref23
ref45
Liu (ref37)
ref42
Naderializadeh (ref36) 2020
ref44
Chou (ref41)
ref43
Iqbal (ref30)
Sutton (ref27) 2018
ref29
ref8
Agarwal (ref21)
Peng (ref33) 2017
ref7
Kingma (ref39) 2014
ref9
ref4
ref3
ref6
ref5
Jiang (ref34)
Jiang (ref20)
ref40
Sukhbaatar (ref32)
Rashid (ref25)
Lillicrap (ref31)
Haarnoja (ref38)
Khan (ref22)
Sunehag (ref24)
References_xml – start-page: 1741
  volume-title: Proc. Int. Conf. Auton. Agents Multiagent Syst. (AAMAS)
  ident: ref21
  article-title: Learning transferable cooperative behavior in multi-agent team
– year: 2017
  ident: ref13
  article-title: Proximal policy optimization algorithms
  publication-title: arXiv:1707.06347
– start-page: 1
  volume-title: Proc. Int. Conf. Learn. Represent. (ICLR)
  ident: ref26
  article-title: QPLEX: Duplex dueling multi-agent Q-learning
– start-page: 590
  volume-title: Proc. 3rd Conf. Robot Learn. (CoRL)
  ident: ref37
  article-title: PIC: Permutation invariant critic for multi-agent deep reinforcement learning
– start-page: 1861
  volume-title: Proc. 35th Int. Conf. Mach. Learn. (ICML)
  ident: ref38
  article-title: Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor
– ident: ref10
  doi: 10.1016/j.automatica.2016.06.024
– ident: ref15
  doi: 10.1109/ICRA46639.2022.9812263
– ident: ref2
  doi: 10.1109/IROS51168.2021.9635898
– year: 2014
  ident: ref39
  article-title: Adam: A method for stochastic optimization
  publication-title: arXiv:1412.6980
– ident: ref44
  doi: 10.1109/IROS.2018.8593871
– ident: ref19
  doi: 10.1109/MSP.2020.3016143
– ident: ref5
  doi: 10.1109/TII.2012.2219061
– year: 2020
  ident: ref36
  article-title: Graph convolutional value decomposition in multi-agent reinforcement learning
  publication-title: arXiv:2010.04740
– start-page: 834
  volume-title: Proc. 34th Int. Conf. Mach. Learn. (ICML)
  ident: ref41
  article-title: Improving stochastic policy gradients in continuous control with deep reinforcement learning using the beta distribution
– ident: ref42
  doi: 10.1109/TAC.2004.834113
– start-page: 25817
  volume-title: Proc. 40th Int. Conf. Mach. Learn. (ICML)
  ident: ref17
  article-title: Scalable multi-agent reinforcement learning through intelligent information aggregation
– ident: ref16
  doi: 10.1109/tnnls.2022.3172168
– year: 2019
  ident: ref46
  article-title: Distillation strategies for proximal policy optimization
  publication-title: arXiv:1901.08128
– ident: ref35
  doi: 10.48550/ARXIV.1706.03762
– start-page: 2961
  volume-title: Proc. 36th Int. Conf. Mach. Learn. (ICML)
  ident: ref30
  article-title: Actor-attention-critic for multi-agent reinforcement learning
– ident: ref29
  doi: 10.1609/aaai.v32i1.11794
– ident: ref8
  doi: 10.1109/tnnls.2021.3104987
– ident: ref3
  doi: 10.1109/LRA.2022.3146912
– ident: ref45
  doi: 10.1109/ICRA46639.2022.9811744
– ident: ref6
  doi: 10.1109/TNNLS.2020.2984944
– start-page: 2244
  volume-title: Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)
  ident: ref32
  article-title: Learning multiagent communication with backpropagation
– start-page: 1
  volume-title: Proc. Int. Conf. Learn. Represent. (ICLR)
  ident: ref31
  article-title: Continuous control with deep reinforcement learning
– start-page: 7254
  volume-title: Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)
  ident: ref34
  article-title: Learning attentional communication for multi-agent cooperation
– ident: ref40
  doi: 10.1109/IROS51168.2021.9635836
– ident: ref1
  doi: 10.1109/OJCOMS.2021.3081996
– volume-title: Reinforcement Learning: An Introduction
  year: 2018
  ident: ref27
– ident: ref7
  doi: 10.1109/TNNLS.2021.3095431
– ident: ref12
  doi: 10.1007/978-3-030-60990-0_12
– start-page: 823
  volume-title: Proc. 3rd Conf. Robot Learn. (CoRL)
  ident: ref22
  article-title: Graph policy gradients for large scale robot control
– ident: ref11
  doi: 10.1109/TAC.2015.2465071
– start-page: 6379
  volume-title: Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)
  ident: ref28
  article-title: Multi-agent actor-critic for mixed cooperative-competitive environments
– start-page: 1
  volume-title: Proc. Int. Conf. Learn. Represent. (ICLR)
  ident: ref20
  article-title: Graph convolutional reinforcement learning
– ident: ref23
  doi: 10.1109/LRA.2021.3074885
– ident: ref9
  doi: 10.1109/TCYB.2020.3000264
– ident: ref14
  doi: 10.48550/arXiv.1812.05905
– start-page: 4295
  volume-title: Proc. 35th Int. Conf. Mach. Learn. (ICML)
  ident: ref25
  article-title: QMIX: Monotonic value function factorisation for deep multi-agent reinforcement learning
– start-page: 2085
  volume-title: Proc. Int. Conf. Auton. Agents MultiAgent Syst. (AAMAS)
  ident: ref24
  article-title: Value-decomposition networks for cooperative multi-agent learning
– ident: ref43
  doi: 10.1109/ICRA.2018.8461113
– ident: ref4
  doi: 10.1109/TSMC.2019.2961753
– year: 2022
  ident: ref18
  article-title: A survey of multi-agent reinforcement learning with communication
  publication-title: arXiv:2203.08975
– year: 2017
  ident: ref33
  article-title: Multiagent bidirectionally-coordinated nets: Emergence of human-level coordination in learning to play StarCraft combat games
  publication-title: arXiv:1703.10069
SSID ssj0000605649
Score 2.5452952
Snippet Learning distributed cooperative policies for large-scale multirobot systems remains a challenging task in the multiagent reinforcement learning (MARL)...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Enrichment Source
Publisher
StartPage 665
SubjectTerms Distributed coordination
graph neural network (GNN)
Multi-robot systems
multiagent reinforcement learning (MARL)
multirobot system
Protocols
Reinforcement learning
Robot kinematics
Scalability
Simulation
soft \text{actor(!{-}!)critic} (SAC) algorithm
Task analysis
Title Graph Soft Actor-Critic Reinforcement Learning for Large-Scale Distributed Multirobot Coordination
URI https://ieeexplore.ieee.org/document/10314538
https://www.ncbi.nlm.nih.gov/pubmed/37948149
https://www.proquest.com/docview/2889243485
Volume 36
WOSCitedRecordID wos001106580100001&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: 2162-2388
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0000605649
  issn: 2162-237X
  databaseCode: RIE
  dateStart: 20120101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT9wwEB4VxKEXaHl1-0BG4lZ5SWzHdo6IFjigVcXuYW-R7TgICW3QPvr7O-MkKy4g9WZFdhTlG9szY8_3AVwIJ5zBqItn0tYcPeKGu0I77kUWo9QWTaxOYhNmMrHzefmnL1ZPtTAxxnT5LI6pmc7y6zZsKFV2SZIECmfoDuwYo7tirW1CJUPHXCd3V-RacCHNfCiSycrL2WRyPx2TVvhYSlEWkiTgpCGuEqLRfLUnJZGVt_3NtO_cHPznF3-C_d7BZFedRXyGD3FxCAeDeAPr5_IR-FuiqmZTXIbZFWXuead6wB5iIlMNKW_Iev7VR4aP2D1dG-dThDWyX8S4S2JZsWapinfZ-nbNrluMZp-6FOMxzG5-z67veC-4wAP6VWteisbXxrtch6Jp8syH4L1T2Gq0cEEVFkNvVxSizmWRORsdOoy4y0bTGFVKeQK7i3YRvwCrc41xI_EtqloplzttdKO8VY10RCk3gnz441XoychJE-O5SkFJVlYJsIoAq3rARvBzO-alo-J4t_cxwfGqZ4fECM4HZCucSXQ84hax3awqYS0Go1LZYgSnHeTb0YOlfH3jrd_goyBh4JSb-Q676-Um_oC98Hf9tFqeobnO7Vky13-qMuPm
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5BQaKXlkcpy9NI3JC3iV9xjlWhFLFEiN3D3iLbcVClalNtd_n9nXGSVS9F4mZFdhTlG9szY8_3AXwSTrgCoy6eSdtw9Ihb7rRx3IssRmksmliTxCaKqrLLZflrKFZPtTAxxnT5LE6pmc7ymy5sKVV2QpIECmfoQ3iklRJZX661S6lk6Jqb5PCK3AguZLEcy2Sy8mRRVbP5lNTCp1KKUksSgZMFsZUQkeadXSnJrNzvcaad5_zwP7_5KRwMLiY77W3iGTyIq-dwOMo3sGE2vwD_jciq2RwXYnZKuXve6x6w3zHRqYaUOWQDA-sfho_YjC6O8zkCG9kX4twluazYsFTHu-58t2FnHcazl32S8QgW518XZxd8kFzgAT2rDS9F65vCu9wE3bZ55kPw3ilstUa4oLTF4NtpLZpc6szZ6NBlxH02Fm2hSilfwt6qW8VXwJrcYORIjIuqUcrlzhSmVd6qVjoilZtAPv7xOgx05KSKcVWnsCQr6wRYTYDVA2AT-Lwbc92Tcfyz9xHBcadnj8QEPo7I1jiX6IDErWK3vamFtRiOSmX1BI57yHejR0t5fc9bP8CTi8XPWT37Xv14A_uCZIJTpuYt7G3W2_gOHoe_m8ub9ftktLdMH-ZF
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=Graph+Soft+Actor%E2%80%93Critic+Reinforcement+Learning+for+Large-Scale+Distributed+Multirobot+Coordination&rft.jtitle=IEEE+transaction+on+neural+networks+and+learning+systems&rft.au=Hu%2C+Yifan&rft.au=Fu%2C+Junjie&rft.au=Wen%2C+Guanghui&rft.date=2025-01-01&rft.issn=2162-237X&rft.eissn=2162-2388&rft.volume=36&rft.issue=1&rft.spage=665&rft.epage=676&rft_id=info:doi/10.1109%2FTNNLS.2023.3329530&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TNNLS_2023_3329530
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2162-237X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2162-237X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2162-237X&client=summon