elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Impressum | Datenschutz | Barrierefreiheit | Kontakt | English
Schriftgröße: [-] Text [+]

SAE level 3 Autonomous driving in urban setting using cognitive architecture

Austel, Benedikt und Weber, Lars und Osterloh, Jan-Patrick (2026) SAE level 3 Autonomous driving in urban setting using cognitive architecture. In: Human Factors and Ergonomics Society Europe Chapter Annual Meeting, HFES 2026. HFES Europe Chapter annual meeting 2026, 2026-04-22 - 2026-04-24, Copenhagen, Denmark. (nicht veröffentlicht)

Dies ist die aktuellste Version dieses Eintrags.

[img] PDF
630kB

Offizielle URL: https://www.hfes-europe.org/wp-content/uploads/2026/04/BenediktAustel2026.pdf

Kurzfassung

The increasing prevalence of automation in the automotive domain necessitates a deeper understanding of these novel human-machine interactions. While SAE level 1 and 2 automation include driving assistance systems, SAE level 3 automation fully drives the vehicle and features the human driver as the fallback. This introduces new tasks such as understanding the automation's capabilities and building situation awareness after previously not paying attention. In order to seamlessly integrate human factors into the design process, it is important to incorporate them in the early design phases. To this end virtual testers can be deployed to interact with early iterations of the model, and allow for a running large quantities of simulations using a large variety of driving styles. This study investigated human performance of 22 participants and a virtual driver model based on the cognitive architecture CASCaS when activating automation takeovers within a driving simulator. The scenario features construction sites blocking the lane, thus requiring a takeover using the lane for the oncoming traffic. The automation can plan and follow the trajectory but relies on the driver to find a gap in the oncoming traffic and initiate the automated takeover maneuver at an opportune time.

Results revealed an increase in error rate with increasing construction site length by human participants, coupled with limited improvement in gap choice and timing over the course of the experiment. Notably, human participants' performance otherwise mirrored that of the virtual driver and the different driving styles could easily be reproduced through changed parameters. These findings highlight the challenges of accurately predicting human behavior in automated driving systems and underscore the potential of virtual driver models like CASCaS for evaluating and adapting to diverse driving styles during the early design phases of Level 3 automation.

elib-URL des Eintrags:https://elib.dlr.de/223164/
Dokumentart:Konferenzbeitrag (Poster)
Titel:SAE level 3 Autonomous driving in urban setting using cognitive architecture
Autoren:
AutorenInstitution oder E-Mail-AdresseAutoren-ORCID-iDORCID Put Code
Austel, Benediktbenedikt.austel (at) dlr.dehttps://orcid.org/0000-0001-7483-8103NICHT SPEZIFIZIERT
Weber, Larslars.weber (at) dlr.deNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Osterloh, Jan-Patrickjan-patrick.osterloh (at) dlr.dehttps://orcid.org/0000-0002-4200-3002NICHT SPEZIFIZIERT
Datum:April 2026
Erschienen in:Human Factors and Ergonomics Society Europe Chapter Annual Meeting, HFES 2026
Referierte Publikation:Nein
Open Access:Ja
Gold Open Access:Nein
In SCOPUS:Nein
In ISI Web of Science:Nein
Status:nicht veröffentlicht
Stichwörter:Cognitive Architecture, Human Centered, HMI, Automotive, Autonomous
Veranstaltungstitel:HFES Europe Chapter annual meeting 2026
Veranstaltungsort:Copenhagen, Denmark
Veranstaltungsart:internationale Konferenz
Veranstaltungsbeginn:22 April 2026
Veranstaltungsende:24 April 2026
Veranstalter :Human Factors and Ergonomics Society, Chapter Europe; hosted by Technical University of Denmark
HGF - Forschungsbereich:Luftfahrt, Raumfahrt und Verkehr
HGF - Programm:Verkehr
HGF - Programmthema:Straßenverkehr
DLR - Schwerpunkt:Verkehr
DLR - Forschungsgebiet:V ST Straßenverkehr
DLR - Teilgebiet (Projekt, Vorhaben):V - V&V4NGC - Methoden, Prozesse und Werkzeugketten für die Validierung & Verifikation von NGC, V - ACT4Transformation - Automated and Connected Technologies for Mobility Transformation
Standort: Oldenburg
Institute & Einrichtungen:Institut für Systems Engineering für zukünftige Mobilität
Hinterlegt von: Austel, Benedikt
Hinterlegt am:10 Mär 2026 11:25
Letzte Änderung:19 Aug 2026 06:32

Verfügbare Versionen dieses Eintrags

  • SAE level 3 Autonomous driving in urban setting using cognitive architecture. (deposited 10 Mär 2026 11:25) [Gegenwärtig angezeigt]

Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags

Blättern
Suchen
Hilfe & Kontakt
Informationen
OpenAIRE Validator logo electronic library verwendet EPrints 3.3.12
Gestaltung Webseite und Datenbank: Copyright © Deutsches Zentrum für Luft- und Raumfahrt (DLR). Alle Rechte vorbehalten.