Flow: A Modular Learning Framework for Mixed Autonomy Traffic

The rapid development of autonomous vehicles (AVs) holds vast potential for transportation systems through improved safety, efficiency, and access to mobility. However, the progression of these impacts, as AVs are adopted, is not well understood. Numerous technical challenges arise from the goal of...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on robotics Ročník 38; číslo 2; s. 1270 - 1286
Hlavní autoři:	Wu, Cathy, Kreidieh, Abdul Rahman, Parvate, Kanaad, Vinitsky, Eugene, Bayen, Alexandre M.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York IEEE 01.04.2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Analytical models Automation technologies for smart cities Autonomy Control theory Deep learning deep learning in robotics and automation deep reinforcement learning Human performance intelligent transportation systems Lane changing Mathematical model Network control Neural networks Numerical models Robots System dynamics Traffic congestion Traffic jams Traffic models Transportation systems Uncertainty Vehicle dynamics
ISSN:	1552-3098, 1941-0468
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	The rapid development of autonomous vehicles (AVs) holds vast potential for transportation systems through improved safety, efficiency, and access to mobility. However, the progression of these impacts, as AVs are adopted, is not well understood. Numerous technical challenges arise from the goal of analyzing the partial adoption of autonomy: partial control and observation, multivehicle interactions, and the sheer variety of scenarios represented by real-world networks. To shed light into near-term AV impacts, this article studies the suitability of deep reinforcement learning (RL) for overcoming these challenges in a low AV-adoption regime. A modular learning framework is presented, which leverages deep RL to address complex traffic dynamics. Modules are composed to capture common traffic phenomena (stop-and-go traffic jams, lane changing, intersections). Learned control laws are found to improve upon human driving performance, in terms of system-level velocity, by up to 57% with only 4-7% adoption of AVs. Furthermore, in single-lane traffic, a small neural network control law with only local observation is found to eliminate stop-and-go traffic-surpassing all known model-based controllers to achieve near-optimal performance-and generalize to out-of-distribution traffic densities.
AbstractList	The rapid development of autonomous vehicles (AVs) holds vast potential for transportation systems through improved safety, efficiency, and access to mobility. However, the progression of these impacts, as AVs are adopted, is not well understood. Numerous technical challenges arise from the goal of analyzing the partial adoption of autonomy: partial control and observation, multivehicle interactions, and the sheer variety of scenarios represented by real-world networks. To shed light into near-term AV impacts, this article studies the suitability of deep reinforcement learning (RL) for overcoming these challenges in a low AV-adoption regime. A modular learning framework is presented, which leverages deep RL to address complex traffic dynamics. Modules are composed to capture common traffic phenomena (stop-and-go traffic jams, lane changing, intersections). Learned control laws are found to improve upon human driving performance, in terms of system-level velocity, by up to 57% with only 4–7% adoption of AVs. Furthermore, in single-lane traffic, a small neural network control law with only local observation is found to eliminate stop-and-go traffic—surpassing all known model-based controllers to achieve near-optimal performance—and generalize to out-of-distribution traffic densities.
Author	Parvate, Kanaad Bayen, Alexandre M. Wu, Cathy Kreidieh, Abdul Rahman Vinitsky, Eugene
Author_xml	– sequence: 1 givenname: Cathy orcidid: 0000-0001-8594-303X surname: Wu fullname: Wu, Cathy email: cathywu@mit.edu organization: Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA, USA – sequence: 2 givenname: Abdul Rahman surname: Kreidieh fullname: Kreidieh, Abdul Rahman email: aboudy@berkeley.edu organization: Department of Civil and Environmental Engineering, University of California, Berkeley, Berkeley, CA, USA – sequence: 3 givenname: Kanaad surname: Parvate fullname: Parvate, Kanaad email: kanaad@berkeley.edu organization: Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA – sequence: 4 givenname: Eugene surname: Vinitsky fullname: Vinitsky, Eugene email: evinitsky@berkeley.edu organization: Department of Mechanical Engineering, University of California, Berkeley, Berkeley, CA, USA – sequence: 5 givenname: Alexandre M. orcidid: 0000-0002-6697-222X surname: Bayen fullname: Bayen, Alexandre M. email: bayen@berkeley.edu organization: Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, USA
BookMark	eNp9kE1LAzEQhoNUsK3eBS8Bz1vztfkQPJRiVWgpSD2HuJ1I6nZTs7vU_nu3tHjw4Gnm8D7vDM8A9apYAULXlIwoJeZu-boYMcLoiBOtOBVnqE-NoBkRUve6Pc9ZxonRF2hQ12tCmDCE99HDtIy7ezzG87hqS5fwDFyqQvWBp8ltYBfTJ_Yx4Xn4hhUet02s4maPl8l5H4pLdO5dWcPVaQ7R2_RxOXnOZounl8l4lhU8V03mvXaGKwVc555pTmUhXPesB8U103KlvVDqXUlHDS2MKDg4oJAzr4RklPIhuj32blP8aqFu7Dq2qepOWiaFEizPpelS8pgqUqzrBN4WoXFNiFWTXCgtJfagynaq7EGVPanqQPIH3KawcWn_H3JzRAIA_MaN0IYTzn8AyN5zrQ
CODEN	ITREAE
CitedBy_id	crossref_primary_10_1109_JIOT_2023_3299934 crossref_primary_10_1109_TASE_2022_3168621 crossref_primary_10_3390_s23104710 crossref_primary_10_1109_TCNS_2023_3338234 crossref_primary_10_1109_TTE_2024_3383091 crossref_primary_10_1109_JIOT_2025_3552554 crossref_primary_10_1109_TMC_2024_3435436 crossref_primary_10_1109_MCS_2024_3503372 crossref_primary_10_1016_j_physa_2024_129991 crossref_primary_10_1109_TIV_2023_3297310 crossref_primary_10_1109_TFUZZ_2024_3450892 crossref_primary_10_3390_s23198229 crossref_primary_10_1109_LRA_2024_3440834 crossref_primary_10_1162_neco_a_01764 crossref_primary_10_1109_ACCESS_2024_3370469 crossref_primary_10_1016_j_icte_2024_09_001 crossref_primary_10_1016_j_trc_2024_104757 crossref_primary_10_1016_j_ijtst_2025_06_002 crossref_primary_10_1049_itr2_12402 crossref_primary_10_1109_TITS_2023_3344473 crossref_primary_10_1109_TITS_2024_3386200 crossref_primary_10_1049_itr2_12492 crossref_primary_10_1109_ACCESS_2024_3430933 crossref_primary_10_1007_s12083_022_01434_0 crossref_primary_10_1109_TMC_2024_3389987 crossref_primary_10_1177_00368504241263406 crossref_primary_10_1109_TITS_2025_3550418 crossref_primary_10_7467_KSAE_2024_32_8_657 crossref_primary_10_1109_TRO_2023_3262120 crossref_primary_10_1016_j_physa_2022_127512 crossref_primary_10_1016_j_trc_2023_104449 crossref_primary_10_1177_02783649231215371 crossref_primary_10_1109_JIOT_2023_3335292 crossref_primary_10_7467_KSAE_2024_32_7_569 crossref_primary_10_1109_TITS_2024_3503092 crossref_primary_10_1080_19427867_2024_2305018 crossref_primary_10_1109_MITS_2022_3174410 crossref_primary_10_1177_03611981251349433 crossref_primary_10_1109_TRO_2024_3463407 crossref_primary_10_1109_MCS_2024_3503402 crossref_primary_10_1177_02783649241284069 crossref_primary_10_1109_JIOT_2024_3481257 crossref_primary_10_1109_JIOT_2023_3303039 crossref_primary_10_1109_TCST_2023_3288636 crossref_primary_10_1109_TITS_2023_3250704 crossref_primary_10_1109_TVT_2023_3312574 crossref_primary_10_1109_TITS_2024_3514117 crossref_primary_10_1177_03611981221144283 crossref_primary_10_1049_itr2_12542 crossref_primary_10_1155_2024_6509852 crossref_primary_10_1109_LCSYS_2024_3499743 crossref_primary_10_1109_TIV_2023_3287131 crossref_primary_10_1109_COMST_2024_3370169 crossref_primary_10_3390_systems11030134 crossref_primary_10_1109_OJITS_2025_3606539 crossref_primary_10_1016_j_trc_2023_104230 crossref_primary_10_1016_j_trc_2023_104274 crossref_primary_10_1109_TRO_2024_3443504
Cites_doi	10.1016/j.trc.2014.04.014 10.1109/TITS.2003.821340 10.1109/70.768189 10.1299/jsmec.43.671 10.1016/j.trd.2009.01.004 10.1016/j.trc.2010.10.004 10.1109/IROS.2012.6386109 10.1109/ITSC.2017.8317694 10.1109/TITS.2003.811644 10.1109/ROBOT.1996.509186 10.1145/504729.504754 10.1109/IVS.2017.7995897 10.1613/jair.2502 10.23919/ACC.2004.1384058 10.1109/TNN.1998.712192 10.1143/JPSJ.71.2339 10.1109/70.75906 10.1002/rob.20147 10.1111/j.2517-6161.1961.tb00391.x 10.1109/87.370698 10.1016/j.trc.2018.02.005 10.1016/0191-2615(84)90019-5 10.1109/ITSC.2018.8569485 10.1109/TRA.2002.807557 10.1512/iumj.1957.6.56038 10.1007/978-3-642-03991-1 10.1109/CDC.1995.478953 10.1109/70.938381 10.1109/TITS.2016.2600504 10.1243/095440705X9407 10.1103/PhysRevE.62.1805 10.1109/JPROC.2003.819610 10.1007/978-3-319-05990-7_20 10.1109/TITS.2003.821292 10.1109/ICRA.2018.8460567 10.1145/3302509.3313784 10.1109/TITS.2017.2725912 10.1109/ITSC.2017.8317733 10.1177/0278364912444766 10.1038/s41586-020-2939-8 10.1007/978-3-319-71682-4_10 10.1137/S0036139997326946 10.1088/1367-2630/10/3/033001 10.1109/TVT.2014.2356204 10.1109/JIOT.2020.2966506 10.1109/TRO.2015.2411371 10.1016/j.trc.2017.02.024 10.1016/s0965-8564(97)88297-3 10.1109/TAC.2019.2921659 10.1145/1015330.1015430 10.1613/jair.3912 10.1115/DETC2011-48829 10.1098/rspa.1955.0089 10.1115/1.2896526 10.1007/BF03167222 10.1109/TVT.2002.800617 10.15607/RSS.2016.XII.029 10.1109/TITS.2016.2587582 10.1287/opre.4.1.42 10.1145/3219819.3220096 10.1103/PhysRevE.51.1035 10.1109/TITS.2014.2345663 10.1016/j.physa.2009.02.033 10.1109/JAS.2016.7508798 10.1137/S0036139997332099 10.1007/978-3-642-32460-4 10.1109/ITSC.2013.6728204 10.1109/TITS.2013.2292500 10.1098/rsta.2010.0205 10.1109/25.260745 10.1111/sapm.12144 10.1016/S0191-2615(00)00042-4 10.1016/j.conengprac.2013.10.008 10.1109/70.88128 10.1007/s12205-016-0382-z 10.1109/ITSC.2018.8569615 10.1109/TITS.2005.844226 10.1145/3357384.3357902 10.1016/j.tra.2015.12.001 10.1016/0191-2615(81)90037-0 10.1142/S0217595907001383 10.1017/9781009002127.013 10.1109/TRO.2009.2011526 10.1016/j.trpro.2017.05.011 10.1007/978-3-540-47641-2_62 10.1016/j.eswa.2011.09.024 10.1109/TITS.2006.884615 10.1177/0278364915581863
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 7TB 8FD FR3 JQ2 L7M L~C L~D
DOI	10.1109/TRO.2021.3087314
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/IET Electronic Library url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1941-0468
EndPage	1286
ExternalDocumentID	10_1109_TRO_2021_3087314 9489303
Genre	orig-research
GrantInformation_xml	– fundername: National Science Foundation Graduate Research Fellowship Program
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-c357t-ff8a9377e385f28316c4a202fe738286d8f477b76a191c94c3eae1e52f7462113
IEDL.DBID	RIE
ISICitedReferencesCount	114
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000732279300001&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	Mon Jun 30 04:05:24 EDT 2025 Sat Nov 29 01:47:28 EST 2025 Tue Nov 18 22:43:55 EST 2025 Wed Aug 27 02:40:50 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	2
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-c357t-ff8a9377e385f28316c4a202fe738286d8f477b76a191c94c3eae1e52f7462113
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0001-8594-303X 0000-0002-6697-222X
PQID	2647425569
PQPubID	27625
PageCount	17
ParticipantIDs	crossref_primary_10_1109_TRO_2021_3087314 proquest_journals_2647425569 crossref_citationtrail_10_1109_TRO_2021_3087314 ieee_primary_9489303
PublicationCentury	2000
PublicationDate	2022-April 2022-4-00 20220401
PublicationDateYYYYMMDD	2022-04-01
PublicationDate_xml	– month: 04 year: 2022 text: 2022-April
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on robotics
PublicationTitleAbbrev	TRO
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	ref57 ref59 ref58 ref53 Corporation (ref41) ref52 ref55 ref54 Mnih (ref80) 2013 Schulman (ref81) 2016 Payne (ref18) 1979 ref51 ref50 ref46 ref48 ref47 ref42 ref44 ref43 Liang (ref103) 2017 ref49 ref8 ref7 ref3 DOT (ref2) 2016 ref6 Buehler (ref4) 2009; 56 ref5 ref100 ref40 Vinitsky (ref110) 2018 ref35 ref34 ref37 ref36 ref31 ref30 ref33 ref32 ref38 Brockman (ref101) 2016 Au (ref56) 2014 ref24 ref23 ref26 ref20 Duan (ref105) 2016 ref22 ref21 ref28 Atkeson (ref11) 1997; 97 ref27 ref29 Shalev-Shwartz (ref10) 2016 Schulman (ref82) 2015 Lefevre (ref39) 2014 ref12 ref15 ref14 Heidemann (ref25) 1996 Heess (ref83) 2015 ref99 ref17 ref16 (ref68) 2007 Bojarski (ref9) 2016 ref19 ref92 Wu (ref93) 2017 ref95 ref94 ref91 ref90 ref89 Levine (ref13) 2016; 17 ref86 ref85 Treiber (ref106) 2004 ref88 ref87 (ref45) 2010 ref84 ref79 ref108 ref78 ref109 ref75 ref104 ref74 ref77 ref76 ref1 Haykin (ref97) 1994 ref71 ref111 ref70 ref112 ref73 Krajzewicz (ref102) 2012; 5 ref72 Sutton (ref96) 2000 Chung (ref98) 2015 ref67 ref69 ref64 ref115 ref63 ref116 ref66 ref113 ref65 ref114 Wu (ref107) 2017; 78 ref60 ref62 ref61
References_xml	– ident: ref76 doi: 10.1016/j.trc.2014.04.014 – ident: ref71 doi: 10.1109/TITS.2003.821340 – ident: ref28 doi: 10.1109/70.768189 – year: 2010 ident: ref45 article-title: Intelligent transport systems, intelligent transport systems - adaptive cruise control systems - performance requirements and test procedures – ident: ref66 doi: 10.1299/jsmec.43.671 – ident: ref69 doi: 10.1016/j.trd.2009.01.004 – ident: ref91 doi: 10.1016/j.trc.2010.10.004 – ident: ref100 doi: 10.1109/IROS.2012.6386109 – ident: ref92 doi: 10.1109/ITSC.2017.8317694 – volume: 97 start-page: 12 volume-title: Proc. Int. Conf. Mach. Learn. year: 1997 ident: ref11 article-title: Robot learning from demonstration – ident: ref40 doi: 10.1109/TITS.2003.811644 – start-page: 2067 volume-title: Proc. Int. Conf. Mach. Learn. year: 2015 ident: ref98 article-title: Gated feedback recurrent neural networks – volume-title: Proc. 12th Int. Symp. Adv. Veh. Control year: 2014 ident: ref39 article-title: Lane keeping assistance with learning-based driver model and model predictive control – ident: ref42 doi: 10.1109/ROBOT.1996.509186 – ident: ref27 doi: 10.1145/504729.504754 – start-page: 399 volume-title: Proc. Conf. Robot Learn. year: 2018 ident: ref110 article-title: Benchmarks for reinforcement learning in mixed-autonomy traffic – volume-title: Popular Sci. Monthly ident: ref41 article-title: Four wheels on jacks park car – ident: ref52 doi: 10.1109/IVS.2017.7995897 – ident: ref6 doi: 10.1613/jair.2502 – volume-title: Proc. 13th Int. Symp. Transp. Traffic Theory year: 1996 ident: ref25 article-title: A queueing theory approach to speed-flow-density relationships – ident: ref50 doi: 10.23919/ACC.2004.1384058 – ident: ref79 doi: 10.1109/TNN.1998.712192 – ident: ref114 doi: 10.1143/JPSJ.71.2339 – ident: ref31 doi: 10.1109/70.75906 – ident: ref3 doi: 10.1002/rob.20147 – ident: ref24 doi: 10.1111/j.2517-6161.1961.tb00391.x – ident: ref35 doi: 10.1109/87.370698 – ident: ref58 doi: 10.1016/j.trc.2018.02.005 – ident: ref112 doi: 10.1016/0191-2615(84)90019-5 – ident: ref108 doi: 10.1109/ITSC.2018.8569485 – ident: ref30 doi: 10.1109/TRA.2002.807557 – ident: ref94 doi: 10.1512/iumj.1957.6.56038 – volume: 56 volume-title: The DARPA Urban Challenge: Autonomous Vehicles in City Traffic year: 2009 ident: ref4 doi: 10.1007/978-3-642-03991-1 – year: 2016 ident: ref10 article-title: Safe, multi-agent, reinforcement learning for autonomous driving – ident: ref78 doi: 10.1109/CDC.1995.478953 – ident: ref29 doi: 10.1109/70.938381 – ident: ref63 doi: 10.1109/TITS.2016.2600504 – ident: ref48 doi: 10.1243/095440705X9407 – ident: ref23 doi: 10.1103/PhysRevE.62.1805 – ident: ref70 doi: 10.1109/JPROC.2003.819610 – year: 2013 ident: ref80 article-title: Playing atari with deep reinforcement learning – ident: ref5 doi: 10.1007/978-3-319-05990-7_20 – year: 2016 ident: ref9 article-title: End to end learning for self-driving cars – ident: ref47 doi: 10.1109/TITS.2003.821292 – year: 2004 ident: ref106 article-title: Multi-anticipative driving in microscopic traffic models – year: 2016 ident: ref2 publication-title: National transportation statistics – ident: ref53 doi: 10.1109/ICRA.2018.8460567 – ident: ref111 doi: 10.1145/3302509.3313784 – ident: ref88 doi: 10.1109/TITS.2017.2725912 – year: 2017 ident: ref93 article-title: Flow: Architecture and benchmarking for reinforcement learning in traffic control – start-page: 1329 volume-title: Int. Conf. Mach. Learning year: 2016 ident: ref105 article-title: Benchmarking deep reinforcement learning for continuous control – ident: ref116 doi: 10.1109/ITSC.2017.8317733 – ident: ref59 doi: 10.1177/0278364912444766 – start-page: 2944 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2015 ident: ref83 article-title: Learning continuous control policies by stochastic value gradients – ident: ref84 doi: 10.1038/s41586-020-2939-8 – ident: ref57 doi: 10.1007/978-3-319-71682-4_10 – volume-title: Proc. Int. Conf. Learn. Representations year: 2016 ident: ref81 article-title: High-dimensional continuous control using generalized advantage estimation – start-page: 1889 volume-title: Proc. Int. Conf. Mach. Learn. year: 2015 ident: ref82 article-title: Trust region policy optimization – ident: ref113 doi: 10.1137/S0036139997326946 – ident: ref14 doi: 10.1088/1367-2630/10/3/033001 – ident: ref38 doi: 10.1109/TVT.2014.2356204 – ident: ref65 doi: 10.1109/JIOT.2020.2966506 – volume: 5 issue: 3&4 year: 2012 ident: ref102 article-title: Recent development and applications of sumo-simulation of urban mobility publication-title: Int. J. Adv. Syst. Meas. – ident: ref32 doi: 10.1109/TRO.2015.2411371 – ident: ref85 doi: 10.1016/j.trc.2017.02.024 – ident: ref74 doi: 10.1016/s0965-8564(97)88297-3 – issue: 722 year: 1979 ident: ref18 article-title: Freflo: A macroscopic simulation model of freeway traffic publication-title: Transp. Res. Rec. – ident: ref44 doi: 10.1109/TAC.2019.2921659 – ident: ref12 doi: 10.1145/1015330.1015430 – ident: ref99 doi: 10.1613/jair.3912 – year: 2016 ident: ref101 article-title: Openai gym – ident: ref75 doi: 10.1115/DETC2011-48829 – volume: 17 start-page: 1 issue: 39 year: 2016 ident: ref13 article-title: End-to-end training of deep visuomotor policies publication-title: J. Mach. Learn. Res. – ident: ref16 doi: 10.1098/rspa.1955.0089 – ident: ref51 doi: 10.1115/1.2896526 – ident: ref21 doi: 10.1007/BF03167222 – ident: ref49 doi: 10.1109/TVT.2002.800617 – ident: ref61 doi: 10.15607/RSS.2016.XII.029 – ident: ref64 doi: 10.1109/TITS.2016.2587582 – ident: ref17 doi: 10.1287/opre.4.1.42 – start-page: 1451 volume-title: Proc. Int. Conf. Auton. Agents Multi-Agent Syst. year: 2014 ident: ref56 article-title: Semi-autonomous intersection management – ident: ref90 doi: 10.1145/3219819.3220096 – start-page: 1057 volume-title: Proc. Adv. Neural Inf. Process. Syst. year: 2000 ident: ref96 article-title: Policy gradient methods for reinforcement learning with function approximation – ident: ref22 doi: 10.1103/PhysRevE.51.1035 – volume-title: Neural Networks: A Comprehensive Foundation year: 1994 ident: ref97 – volume: 78 start-page: 398 volume-title: Proc. 1st Annu Conf. Robot Learn. year: 2017 ident: ref107 article-title: Emergent behaviors in mixed-autonomy traffic – year: 2007 ident: ref68 article-title: Internal project on eco-driving in category b driver training & the driving test – ident: ref86 doi: 10.1109/TITS.2014.2345663 – ident: ref55 doi: 10.1016/j.physa.2009.02.033 – ident: ref87 doi: 10.1109/JAS.2016.7508798 – ident: ref19 doi: 10.1137/S0036139997332099 – year: 2017 ident: ref103 article-title: Ray rllib: A composable and scalable reinforcement learning library – ident: ref67 doi: 10.1007/978-3-642-32460-4 – ident: ref7 doi: 10.1109/ITSC.2013.6728204 – ident: ref73 doi: 10.1109/TITS.2013.2292500 – ident: ref15 doi: 10.1098/rsta.2010.0205 – ident: ref8 doi: 10.1109/TAC.2019.2921659 – ident: ref46 doi: 10.1109/25.260745 – ident: ref77 doi: 10.1111/sapm.12144 – ident: ref115 doi: 10.1016/S0191-2615(00)00042-4 – ident: ref37 doi: 10.1016/j.conengprac.2013.10.008 – ident: ref34 doi: 10.1109/70.88128 – ident: ref62 doi: 10.1007/s12205-016-0382-z – ident: ref109 doi: 10.1109/ITSC.2018.8569615 – ident: ref72 doi: 10.1109/TITS.2005.844226 – ident: ref89 doi: 10.1145/3357384.3357902 – ident: ref1 doi: 10.1016/j.tra.2015.12.001 – ident: ref20 doi: 10.1016/0191-2615(81)90037-0 – ident: ref26 doi: 10.1142/S0217595907001383 – ident: ref95 doi: 10.1017/9781009002127.013 – ident: ref33 doi: 10.1109/TRO.2009.2011526 – ident: ref104 doi: 10.1016/j.trpro.2017.05.011 – ident: ref54 doi: 10.1007/978-3-540-47641-2_62 – ident: ref43 doi: 10.1016/j.eswa.2011.09.024 – ident: ref36 doi: 10.1109/TITS.2006.884615 – ident: ref60 doi: 10.1177/0278364915581863
SSID	ssj0024903
Score	2.6653328
Snippet	The rapid development of autonomous vehicles (AVs) holds vast potential for transportation systems through improved safety, efficiency, and access to mobility....
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1270
SubjectTerms	Analytical models Automation technologies for smart cities Autonomy Control theory Deep learning deep learning in robotics and automation deep reinforcement learning Human performance intelligent transportation systems Lane changing Mathematical model Network control Neural networks Numerical models Robots System dynamics Traffic congestion Traffic jams Traffic models Transportation systems Uncertainty Vehicle dynamics
Title	Flow: A Modular Learning Framework for Mixed Autonomy Traffic
URI	https://ieeexplore.ieee.org/document/9489303 https://www.proquest.com/docview/2647425569
Volume	38
WOSCitedRecordID	wos000732279300001&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/IET Electronic Library 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/eLvHCXMwlV3PS8MwFH7M4UEP_pridEoOXgSzdU3aNN6GOLxsikzYrWTpiwiyytapf75J1k5FEbz1kJT2e23zvua97wM4Y5xHQgpNkxgZ5SrmVKLOKONKcR2YCVOBN5sQw2EyHsu7GlysemEQ0RefYdsd-r38LNcL96usI51SipP2XBNCLHu1PnX1pHdBdopilAUyqbYkA9kZ3d9aIhh22079jnX5tyXIe6r8-BD71aW__b_r2oGtMoskvWXYd6GG0z3Y_KIt2AAbkfztkvTIIM9cqSkplVQfSb-qxyI2YSWDp3fMSG9R-O4GYtcuJyqxDw_969HVDS29EqhmkSioMYmymYZAlkTGpgzdWHNlb9qgYK5TPEsMF2IiYmUJmpZcM1TYxSg0gseWBLIDqE_zKR4CYZKpMEZpJia251CW8zCjnSm5yrjOoiZ0KvhSXQqJOz-L59QTikCmFvDUAZ6WgDfhfDXjZSmi8cfYhgN4Na7EtgmtKkJp-ZbNU5vMWWYfRbE8-n3WMWyErl3BV9q0oF7MFngC6_q1eJrPTv0D9AGimcDG
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3PS8MwFH6MKagHf01xOjUHL4LZ2iZtGm9DHBO3KTJht5KliQxkla1T_3yTrJuKInjrISnt99rmfc173wdwRigNGWcSx5EimIqIYq5kigkVgkpPD4nwnNkE6_XiwYDfl-Bi2QujlHLFZ6puD91efprJmf1V1uBWKcVKe66ElAb-vFvrU1mPOx9kqymGicfjxaakxxv9hztDBQO_bvXviE-_LULOVeXHp9itL62t_13ZNmwWeSRqzgO_AyU13oWNL-qCFTAxyd4uURN1s9QWm6JCS_UJtRYVWcikrKg7elcpas5y19-AzOplZSX24LF13b9q48ItAUsSshxrHQuTazBF4lCbpMGPJBXmprVixPaKp7GmjA1ZJAxFk5xKooTyVRhoRiNDA8k-lMfZWB0AIpyIIFJcD3VkziEM6yFaWltykVKZhlVoLOBLZCElbh0tnhNHKTyeGMATC3hSAF6F8-WMl7mMxh9jKxbg5bgC2yrUFhFKivdsmph0znD7MIz44e-zTmGt3e92ks5N7_YI1gPbvODqbmpQziczdQyr8jUfTScn7mH6ANhyxA0
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=Flow%3A+A+Modular+Learning+Framework+for+Mixed+Autonomy+Traffic&rft.jtitle=IEEE+transactions+on+robotics&rft.au=Wu%2C+Cathy&rft.au=Kreidieh%2C+Abdul+Rahman&rft.au=Parvate%2C+Kanaad&rft.au=Vinitsky%2C+Eugene&rft.date=2022-04-01&rft.issn=1552-3098&rft.eissn=1941-0468&rft.volume=38&rft.issue=2&rft.spage=1270&rft.epage=1286&rft_id=info:doi/10.1109%2FTRO.2021.3087314&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TRO_2021_3087314
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