Show simple item record

AuthorKlimke, Marvindc.contributor.author
AuthorVölz, Benjamindc.contributor.author
AuthorBuchholz, Michaeldc.contributor.author
Date of accession2022-09-26T07:31:19Zdc.date.accessioned
Available in OPARU since2022-09-26T07:31:19Zdc.date.available
Date of first publication2022-07-19dc.date.issued
AbstractUrban intersections are prone to delays and inefficiencies due to static precedence rules and occlusions limiting the view on prioritized traffic. Existing approaches to improve traffic flow, widely known as automatic intersection management systems, are mostly based on non-learning reservation schemes or optimization algorithms. Machine learning-based techniques show promising results in planning for a single ego vehicle. This work proposes to leverage machine learning algorithms to optimize traffic flow at urban intersections by jointly planning for multiple vehicles. Learning-based behavior planning poses several challenges, demanding for a suited input and output representation as well as large amounts of ground-truth data. We address the former issue by using a flexible graph-based input representation accompanied by a graph neural network. This allows to efficiently encode the scene and inherently provide individual outputs for all involved vehicles. To learn a sensible policy, without relying on the imitation of expert demonstrations, the cooperative planning task is considered as a reinforcement learning problem. We train and evaluate the proposed method in an open-source simulation environment for decision making in automated driving. Compared to a first-in-first-out scheme and traffic governed by static priority rules, the learned planner shows a significant gain in flow rate, while reducing the number of induced stops. In addition to synthetic simulations, the approach is also evaluated based on real-world traffic data taken from the publicly available inD dataset.dc.description.abstract
Languageendc.language.iso
PublisherUniversität Ulmdc.publisher
LicenseLizenz Adc.rights
Link to license texthttps://oparu.uni-ulm.de/xmlui/licenseA_v1dc.rights.uri
KeywordKooperative Planungdc.subject
KeywordAutomatisiertes Fahrendc.subject
KeywordVernetzte Fahrzeugedc.subject
KeywordVehicle-to-Infrastructuredc.subject
KeywordGraph Neural Networkdc.subject
Dewey Decimal GroupDDC 620 / Engineering & allied operationsdc.subject.ddc
LCSHAutomated vehiclesdc.subject.lcsh
LCSHReinforcement learningdc.subject.lcsh
TitleCooperative Behavior Planning for Automated Driving Using Graph Neural Networksdc.title
Resource typeBeitrag zu einer Konferenzdc.type
VersionacceptedVersiondc.description.version
DOIhttp://dx.doi.org/10.18725/OPARU-44794dc.identifier.doi
URNhttp://nbn-resolving.de/urn:nbn:de:bsz:289-oparu-44870-7dc.identifier.urn
GNDAutonomes Fahrzeugdc.subject.gnd
GNDNeuronales Netzdc.subject.gnd
FacultyFakultät für Ingenieurwissenschaften, Informatik und Psychologieuulm.affiliationGeneral
InstitutionInstitut für Mess-, Regel- und Mikrotechnikuulm.affiliationSpecific
Peer reviewjauulm.peerReview
DCMI TypeTextuulm.typeDCMI
CategoryPublikationenuulm.category
DOI of original publication10.1109/IV51971.2022.9827230dc.relation1.doi
Source - Title of source2022 IEEE Intelligent Vehicles Symposium (IV)source.title
Source - Place of publicationInstitute of Electrical and Electronics Engineers (IEEE)source.publisher
Source - Volume2022source.volume
Source - Year2022source.year
Source - From page167source.fromPage
Source - To page174source.toPage
Source - ISBN978-1-6654-8821-1source.identifier.isbn
Source - ISBN978-1-6654-8822-8source.identifier.isbn
Conference nameIEEE Intelligent Vehicles Symposiumuulm.conferenceName
Conference placeAachenuulm.conferencePlace
Conference start date2022-06-04uulm.conferenceStartDate
Conference end date2022-06-09uulm.conferenceEndDate
Open AccessGreen Accepted, Green Publisheduulm.OA
WoS000854106700024uulm.identifier.wos
Bibliographyuulmuulm.bibliographie
Project uulmLUKAS / Verbundprojekt: LUKAS - Lokales Umfeldmodell für das kooperative, automatisierte Fahren in komplexen Verkehrssituationen; Teilvorhaben: Infrastrukturseite Datenverarbeitung und kooperative Handlungsplanung / BMWi / 19A20004Fuulm.projectOther


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record