Parallel Multistep Evaluation With Efficient Data Utilization for Safe Neural Critic Control and Its Application to Orbital Maneuver Systems

Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including safety, efficiency, and error accumulation. To address the neglects in safe neural critic control, this article introduces a parallel multis...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE transaction on neural networks and learning systems Vol. 36; no. 9; pp. 17114 - 17127
Main Authors:	Wang, Jiangyu, Wang, Ding, Ren, Jin, Liu, Derong, Qiao, Junfei
Format:	Journal Article
Language:	English
Published:	United States IEEE 01.09.2025
Subjects:	Adaptation models Adaptive dynamic programming (ADP) Aerospace electronics approximate errors Artificial neural networks Cost function Costs Data models data-driven Heuristic algorithms learning systems neural networks (NNs) Optimal control Q-learning safe optimal control Safety
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	Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including safety, efficiency, and error accumulation. To address the neglects in safe neural critic control, this article introduces a parallel multistep evaluation mechanism that combines data from the system interaction with data generated by data-driven models. Based on this evaluation mechanism, we propose a novel parallel multistep Q-learning algorithm that enhances data utilization efficiency and mitigates the error accumulation. Furthermore, we formulate a novel control barrier function (CBF) to ensure safety during learning and control processes, which is capable of dealing with asymmetric constraints and adjusting the constraint strength. In addition, the analysis reveals that multistep information introduced by data-driven models influences the learning performance of actor-critic neural networks (NNs). Finally, parallel multistep Q-learning, which makes use of data in aspects of safety, efficiency, and error bounds, is validated within an orbital maneuver system.
AbstractList	Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including safety, efficiency, and error accumulation. To address the neglects in safe neural critic control, this article introduces a parallel multistep evaluation mechanism that combines data from the system interaction with data generated by data-driven models. Based on this evaluation mechanism, we propose a novel parallel multistep Q-learning algorithm that enhances data utilization efficiency and mitigates the error accumulation. Furthermore, we formulate a novel control barrier function (CBF) to ensure safety during learning and control processes, which is capable of dealing with asymmetric constraints and adjusting the constraint strength. In addition, the analysis reveals that multistep information introduced by data-driven models influences the learning performance of actor-critic neural networks (NNs). Finally, parallel multistep Q-learning, which makes use of data in aspects of safety, efficiency, and error bounds, is validated within an orbital maneuver system.Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including safety, efficiency, and error accumulation. To address the neglects in safe neural critic control, this article introduces a parallel multistep evaluation mechanism that combines data from the system interaction with data generated by data-driven models. Based on this evaluation mechanism, we propose a novel parallel multistep Q-learning algorithm that enhances data utilization efficiency and mitigates the error accumulation. Furthermore, we formulate a novel control barrier function (CBF) to ensure safety during learning and control processes, which is capable of dealing with asymmetric constraints and adjusting the constraint strength. In addition, the analysis reveals that multistep information introduced by data-driven models influences the learning performance of actor-critic neural networks (NNs). Finally, parallel multistep Q-learning, which makes use of data in aspects of safety, efficiency, and error bounds, is validated within an orbital maneuver system. Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including safety, efficiency, and error accumulation. To address the neglects in safe neural critic control, this article introduces a parallel multistep evaluation mechanism that combines data from the system interaction with data generated by data-driven models. Based on this evaluation mechanism, we propose a novel parallel multistep Q-learning algorithm that enhances data utilization efficiency and mitigates the error accumulation. Furthermore, we formulate a novel control barrier function (CBF) to ensure safety during learning and control processes, which is capable of dealing with asymmetric constraints and adjusting the constraint strength. In addition, the analysis reveals that multistep information introduced by data-driven models influences the learning performance of actor-critic neural networks (NNs). Finally, parallel multistep Q-learning, which makes use of data in aspects of safety, efficiency, and error bounds, is validated within an orbital maneuver system.
Author	Qiao, Junfei Wang, Ding Wang, Jiangyu Ren, Jin Liu, Derong
Author_xml	– sequence: 1 givenname: Jiangyu orcidid: 0000-0002-0309-8109 surname: Wang fullname: Wang, Jiangyu email: wangjiangyu@emails.bjut.edu.cn organization: School of Information Science and Technology, Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing Laboratory of Smart Environmental Protection, and Beijing Institute of Artificial Intelligence, Beijing University of Technology, Beijing, China – sequence: 2 givenname: Ding orcidid: 0000-0002-7149-5712 surname: Wang fullname: Wang, Ding email: dingwang@bjut.edu.cn organization: School of Information Science and Technology, Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing Laboratory of Smart Environmental Protection, and Beijing Institute of Artificial Intelligence, Beijing University of Technology, Beijing, China – sequence: 3 givenname: Jin surname: Ren fullname: Ren, Jin email: renjin@emails.bjut.edu.cn organization: School of Information Science and Technology, Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing Laboratory of Smart Environmental Protection, and Beijing Institute of Artificial Intelligence, Beijing University of Technology, Beijing, China – sequence: 4 givenname: Derong orcidid: 0000-0003-3715-4778 surname: Liu fullname: Liu, Derong email: liudr@sustech.edu.cn organization: School of Automation and Intelligent Manufacturing, Southern University of Science and Technology, Shenzhen, China – sequence: 5 givenname: Junfei orcidid: 0000-0003-0950-7900 surname: Qiao fullname: Qiao, Junfei email: adqiao@bjut.edu.cn organization: School of Information Science and Technology, Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing Laboratory of Smart Environmental Protection, and Beijing Institute of Artificial Intelligence, Beijing University of Technology, Beijing, China
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/40440136$$D View this record in MEDLINE/PubMed
BookMark	eNpFkd1OGzEQhS1EBZTyAqhCvuQmqX829u4lCqFFCqFSqNq7le0dC1fOerG9SPQZ-tA13ZTOzYzk75yx5rxHh33oAaFzSuaUkubTw2az3s4ZYYs5X0giqThAJ4wKNmO8rg_fZvnjGJ2l9JOUEmQhquYIHVekqgjl4gT9_qqi8h48vht9dinDgFfPyo8qu9Dj7y4_4pW1zjjoM75WWeFv2Xn3a3q3IeKtsoA3MBYfvIwuO4OXoc8xeKz6Dt_mhK-GwTszSXLA91G7XOg71cP4DMXipSzepQ_onVU-wdm-n6Ltzeph-WW2vv98u7xazwyTMs-4JUbyjtRGLLQhghvaKLBaC2C1ZtpQrnnHKDWUiK5STFva1VYyS2pJ-Sm6nFyHGJ5GSLnduWTA-_KbMKaWM1qxphGyLujFHh31Drp2iG6n4kv7734FYBNgYkgpgn1DKGlfc2r_5tS-5tTucyqij5PIAcB_ASVUStHwPwRNkEU
CODEN	ITNNAL
Cites_doi	10.2514/1.G001154 10.1109/TNNLS.2017.2661865 10.1109/TCBB.2018.2830357 10.1109/TASE.2023.3294187 10.1109/TCYB.2016.2623859 10.1109/TCYB.2021.3054626 10.23919/cje.2024.00.287 10.1109/TSMC.2020.3042876 10.1109/TCSII.2023.3343375 10.1109/TAC.2016.2638961 10.1109/TCYB.2022.3163816 10.1016/j.ins.2024.121640 10.1109/TSMC.2013.2295351 10.1016/j.procs.2022.01.020 10.1109/TNNLS.2013.2280013 10.1016/j.ins.2017.05.005 10.1109/TAC.2018.2790260 10.1109/TNNLS.2021.3135405 10.1109/JAS.2022.105692 10.1109/TCST.2017.2789191 10.1109/JAS.2023.123684 10.1109/TNNLS.2022.3143527 10.1109/TNNLS.2022.3213825 10.1109/TNNLS.2013.2281663 10.1109/TSMC.2017.2771516 10.1016/j.asoc.2024.111687 10.1109/TCYB.2025.3530951 10.1109/TNNLS.2022.3186528 10.1109/JAS.2024.124509 10.23919/cje.2023.00.288 10.1109/TCYB.2022.3232599 10.1109/TIE.2017.2772162 10.1109/TSMCB.2012.2216523 10.1109/TCYB.2014.2354377 10.1109/TAC.2019.2926167 10.1109/TIM.2023.3325870 10.1109/TNNLS.2020.3009015 10.1109/TNNLS.2020.2967871 10.1002/rnc.5132 10.1016/j.neucom.2024.128837 10.1109/TAC.2020.3024161 10.1109/TCYB.2021.3108034 10.1201/9781003143444 10.1109/TCYB.2021.3107801 10.22190/FUME240914044P 10.1109/LRA.2021.3070252 10.1109/TCYB.2016.2542923 10.1016/j.neunet.2021.08.025 10.1109/TNNLS.2022.3178746 10.1109/TNNLS.2023.3326397 10.1007/978-3-319-50815-3 10.1109/JAS.2023.123843 10.1109/TCYB.2022.3233593 10.1109/TRO.2019.2920206 10.1016/j.automatica.2025.112168 10.1109/TNNLS.2022.3152268 10.1109/TCYB.2015.2417170 10.1109/TNNLS.2017.2751018
ContentType	Journal Article
DBID	97E RIA RIE AAYXX CITATION NPM 7X8
DOI	10.1109/TNNLS.2025.3570716
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	MEDLINE - Academic PubMed
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	17127
ExternalDocumentID	40440136 10_1109_TNNLS_2025_3570716 11017769
Genre	orig-research Journal Article
GrantInformation_xml	– fundername: National Natural Science Foundation of China grantid: 62222301; 62473012; 62021003 funderid: 10.13039/501100001809 – fundername: Beijing Natural Science Foundation grantid: F251019 – fundername: National Science and Technology Major Project grantid: 2021ZD0112302; 2021ZD0112301 funderid: 10.13039/501100001809
GroupedDBID	0R~ 4.4 5VS 6IK 97E AAJGR AASAJ AAWTH 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 AARMG ABAZT NPM 7X8
ID	FETCH-LOGICAL-c277t-3f0c73d08c65bc063c19aefbb6e28b2bc13b3d211c106d4a2bf1d8f72f08713
IEDL.DBID	RIE
ISICitedReferencesCount	2
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001499469800001&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 Nov 01 14:40:44 EDT 2025 Sat Sep 06 11:16:10 EDT 2025 Sat Nov 29 07:36:10 EST 2025 Mon Dec 08 03:37:11 EST 2025
IsPeerReviewed	false
IsScholarly	true
Issue	9
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-c277t-3f0c73d08c65bc063c19aefbb6e28b2bc13b3d211c106d4a2bf1d8f72f08713
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ORCID	0000-0003-3715-4778 0000-0003-0950-7900 0000-0002-0309-8109 0000-0002-7149-5712
PMID	40440136
PQID	3214299678
PQPubID	23479
PageCount	14
ParticipantIDs	proquest_miscellaneous_3214299678 pubmed_primary_40440136 ieee_primary_11017769 crossref_primary_10_1109_TNNLS_2025_3570716
PublicationCentury	2000
PublicationDate	2025-09-01
PublicationDateYYYYMMDD	2025-09-01
PublicationDate_xml	– month: 09 year: 2025 text: 2025-09-01 day: 01
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	ref13 ref57 ref12 ref56 ref15 ref14 ref58 ref53 ref52 ref11 ref55 ref10 ref54 ref17 ref16 ref19 ref18 ref51 ref50 ref46 ref45 ref48 ref47 ref42 ref41 ref44 ref43 ref49 ref8 ref7 ref9 ref4 ref3 ref6 ref5 ref40 ref35 ref34 ref37 ref36 ref31 ref30 ref33 ref32 ref2 ref1 ref39 ref38 ref24 ref23 ref26 ref25 ref20 ref22 ref21 ref28 ref27 ref29
References_xml	– ident: ref58 doi: 10.2514/1.G001154 – ident: ref46 doi: 10.1109/TNNLS.2017.2661865 – ident: ref16 doi: 10.1109/TCBB.2018.2830357 – ident: ref15 doi: 10.1109/TASE.2023.3294187 – ident: ref33 doi: 10.1109/TCYB.2016.2623859 – ident: ref50 doi: 10.1109/TCYB.2021.3054626 – ident: ref3 doi: 10.23919/cje.2024.00.287 – ident: ref23 doi: 10.1109/TSMC.2020.3042876 – ident: ref41 doi: 10.1109/TCSII.2023.3343375 – ident: ref55 doi: 10.1109/TAC.2016.2638961 – ident: ref1 doi: 10.1109/TCYB.2022.3163816 – ident: ref7 doi: 10.1016/j.ins.2024.121640 – ident: ref28 doi: 10.1109/TSMC.2013.2295351 – ident: ref9 doi: 10.1016/j.procs.2022.01.020 – ident: ref17 doi: 10.1109/TNNLS.2013.2280013 – ident: ref37 doi: 10.1016/j.ins.2017.05.005 – ident: ref44 doi: 10.1109/TAC.2018.2790260 – ident: ref31 doi: 10.1109/TNNLS.2021.3135405 – ident: ref21 doi: 10.1109/JAS.2022.105692 – ident: ref11 doi: 10.1109/TCST.2017.2789191 – ident: ref40 doi: 10.1109/JAS.2023.123684 – ident: ref57 doi: 10.1109/TNNLS.2022.3143527 – ident: ref6 doi: 10.1109/TNNLS.2022.3213825 – ident: ref49 doi: 10.1109/TNNLS.2013.2281663 – ident: ref38 doi: 10.1109/TSMC.2017.2771516 – ident: ref8 doi: 10.1016/j.asoc.2024.111687 – ident: ref36 doi: 10.1109/TCYB.2025.3530951 – ident: ref2 doi: 10.1109/TNNLS.2022.3186528 – ident: ref12 doi: 10.1109/JAS.2024.124509 – ident: ref5 doi: 10.23919/cje.2023.00.288 – ident: ref47 doi: 10.1109/TCYB.2022.3232599 – ident: ref39 doi: 10.1109/TIE.2017.2772162 – ident: ref48 doi: 10.1109/TSMCB.2012.2216523 – ident: ref43 doi: 10.1109/TCYB.2014.2354377 – ident: ref51 doi: 10.1109/TAC.2019.2926167 – ident: ref29 doi: 10.1109/TIM.2023.3325870 – ident: ref18 doi: 10.1109/TNNLS.2020.3009015 – ident: ref53 doi: 10.1109/TNNLS.2020.2967871 – ident: ref56 doi: 10.1002/rnc.5132 – ident: ref14 doi: 10.1016/j.neucom.2024.128837 – ident: ref13 doi: 10.1109/TAC.2020.3024161 – ident: ref45 doi: 10.1109/TCYB.2021.3108034 – ident: ref27 doi: 10.1201/9781003143444 – ident: ref52 doi: 10.1109/TCYB.2021.3107801 – ident: ref10 doi: 10.22190/FUME240914044P – ident: ref26 doi: 10.1109/LRA.2021.3070252 – ident: ref32 doi: 10.1109/TCYB.2016.2542923 – ident: ref30 doi: 10.1016/j.neunet.2021.08.025 – ident: ref35 doi: 10.1109/TNNLS.2022.3178746 – ident: ref54 doi: 10.1109/TNNLS.2023.3326397 – ident: ref22 doi: 10.1007/978-3-319-50815-3 – ident: ref24 doi: 10.1109/JAS.2023.123843 – ident: ref20 doi: 10.1109/TCYB.2022.3233593 – ident: ref25 doi: 10.1109/TRO.2019.2920206 – ident: ref42 doi: 10.1016/j.automatica.2025.112168 – ident: ref19 doi: 10.1109/TNNLS.2022.3152268 – ident: ref4 doi: 10.1109/TCYB.2015.2417170 – ident: ref34 doi: 10.1109/TNNLS.2017.2751018
SSID	ssj0000605649
Score	2.4904306
Snippet	Data-driven methods have significantly advanced optimal learning control, but some approaches overlook systematic considerations of data utilization, including...
SourceID	proquest pubmed crossref ieee
SourceType	Aggregation Database Index Database Publisher
StartPage	17114
SubjectTerms	Adaptation models Adaptive dynamic programming (ADP) Aerospace electronics approximate errors Artificial neural networks Cost function Costs Data models data-driven Heuristic algorithms learning systems neural networks (NNs) Optimal control Q-learning safe optimal control Safety
Title	Parallel Multistep Evaluation With Efficient Data Utilization for Safe Neural Critic Control and Its Application to Orbital Maneuver Systems
URI	https://ieeexplore.ieee.org/document/11017769 https://www.ncbi.nlm.nih.gov/pubmed/40440136 https://www.proquest.com/docview/3214299678
Volume	36
WOSCitedRecordID	wos001499469800001&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/eLvHCXMwlV1NT9wwEB0B4tBLaQst2w80SL2hQGJnY-eI6KJWgi3SUnVvkT8mAgll0ZLlV_CjO3aS3ROHnpJDHDt5Y80b2zMP4Lska6wQPjEkZJIXpBOtiyxRmTU101OV26hacqWmUz2flzd9snrMhSGiePiMTsNt3Mv3C7cKS2VnWbAfVZTbsM2XLllrvaCSMjEvIt0VWSESIdV8SJJJy7Pb6fRqxuGgGJ_KsWK_GrSL8qC3nMXqzBufFEVWXueb0e9c7v3niN_B255g4nlnEe9hi5oPsDeIN2A_l_fh5cYsg47KA8YcXAb7ESfr0t_49769w0msL8E94A_TGvzT3j_0aZvIXBdnpiYM1T24v04yAS-6k-9oGo-_2ic83-yPY7vA30sbRErw2jS0eg7D6QqmH8DscnJ78TPppRkSJ5RqE1mnTkmfaleMrWOa47LSUG1tQUJbYV0mrfQcXDoOOX1uhK0zr2sl6pQjNPkRdppFQ4eAXomUm5TSKZ1b4rbekVUyp1qXRpkRnAzQVI9d_Y0qxi1pWUVMq4Bp1WM6goOAwebJ_veP4HiAs-LpE_ZE-DsXq6cq6DSxR2aXPYJPHc7r1oN5fH7lrV_gTei8O3H2FXba5Yq-wa57ZtCWR2yjc30UbfQf4hXjXg
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NT9wwEB3xUYleCi203RboVOqtCiR2NnaOCBaBWFKk3ap7i_wVFQll0ZLlV_RHd-wkuycO3HKIY8dvrHlje-YB_OBOK82YjZRjPEozJyMpsyQSiVYV0VOR6qBaMhZFIWez_K5LVg-5MM65cPnMnfjHcJZv52bpt8pOE28_Iss3YXuYpixu07VWWyoxUfMsEF6WZCxiXMz6NJk4P50WxXhCASEbnvChIM_q1YtSr7ichPrMa68UZFZeZpzB81zuvnLMe_Cuo5h41trEe9hw9QfY7eUbsFvN-_DvTi28ksoDhixcgvsRR6vi3_jnvvmLo1BhgnrAC9Uo_N3cP3SJm0hsFyeqcujre1B_rWgCnrd331HVFq-bJzxbn5BjM8dfC-1lSvBW1W757IfTlkw_gMnlaHp-FXXiDJFhQjQRr2IjuI2lyYbaENExSa5cpXXmmNRMm4Rrbim8NBR02lQxXSVWVoJVMcVo_CNs1fPafQa0gsXUJOdGyFQ7amuN04KnrpK5EmoAP3toyse2AkcZIpc4LwOmpce07DAdwIHHYP1mN_0D-N7DWdIC8qci9J_z5VPplZrIJ5PTHsCnFudV6948vrzw1W-wczW9HZfj6-LmK7z1A2nvnx3CVrNYuiN4Y54JwMVxsNT_fU_lvQ
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=Parallel+Multistep+Evaluation+With+Efficient+Data+Utilization+for+Safe+Neural+Critic+Control+and+Its+Application+to+Orbital+Maneuver+Systems&rft.jtitle=IEEE+transaction+on+neural+networks+and+learning+systems&rft.au=Wang%2C+Jiangyu&rft.au=Wang%2C+Ding&rft.au=Ren%2C+Jin&rft.au=Liu%2C+Derong&rft.date=2025-09-01&rft.issn=2162-237X&rft.eissn=2162-2388&rft.volume=36&rft.issue=9&rft.spage=17114&rft.epage=17127&rft_id=info:doi/10.1109%2FTNNLS.2025.3570716&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TNNLS_2025_3570716
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