Nearly half of new vehicle production is expected to be autonomous, SAE International automation Level 5, by 2045, potentially improving safety, lessening traffic congestion, and lowering driver stress. However, many of these benefits will begin to be realized when vehicles reach SAE International Level 3, in which the occupant is allowed to perform secondary tasks instead of solely focusing on the task of driving. The ability to be engaged in non-driving tasks places critical importance on transferring the occupant’s focus back to having complete control of the vehicle. Failure during a vital takeover situation can result in damages, injury, or even death. Therefore, it is imperative to understand what factors influence successful Level 3 vehicle takeovers. Our study investigates whether a driver’s physiological data can help predict takeover performance. To accomplish this, we developed an SAE International Level 3 driving takeover simulation on a driving simulator. The simulation tracked the participant’s interaction with the vehicle after receiving the takeover request (e.g., takeover reaction time, takeover success, and steering rotation), while physiological sensors tracked the participant’s bodily responses (e.g., brain activity, skin conductivity, and heart rate). We have found potential physiological markers that may be used to develop a personalized takeover performance model. Further data analysis and testing will clarify what real-time physiological data results in the best takeover performance.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Exploring Driver Physiological Response During Level 3 Conditional Driving Automation


    Beteiligte:
    Gluck, Aaron (Autor:in) / Deng, Min (Autor:in) / Zhao, Yijin (Autor:in) / Menassa, Carol (Autor:in) / Li, Da (Autor:in) / Brinkley, Julian (Autor:in) / Kamat, Vineet (Autor:in)


    Erscheinungsdatum :

    2022-11-17


    Format / Umfang :

    550798 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch