Perceived Comfort and Safety in Automated Driving based on Physiological Signals
Findings from a Proving Ground Study
J.A. Scharringa (TU Delft - Mechanical Engineering)
Konstantinos Gkentsidis – Mentor (Siemens Digital Industries Software )
R Happee – Graduation committee member (TU Delft - Intelligent Vehicles)
Georgios Papaioannou – Graduation committee member (TU Delft - Intelligent Vehicles)
V. Kotian – Graduation committee member
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
With the advancements in automated driving, there is an increasing focus on passenger comfort and safety, fueled by the desire to establish a unique user experience identity among automotive companies. This study investigates the potential of the Galvanic Skin Response (GSR) as a physiological marker for assessing user experience. For this purpose, a test study of 32 participants was performed by collecting GSR measurements and self-reported comfort and safety scores, using a Wizard-ofOz setup on a closed test-track over repeated laps, alternating between distinct driving styles. Statistical analysis revealed the phasic maximum amplitude and peak count as the features most strongly correlating with both objective driving style and perceived comfort and safety ratings. The GSR measurements were also given as input for a predictive model for classifying the driving style, yielding an accuracy of 88.61%. General performance of the same model for perceived comfort and safety prediction on a five-point Likert scale was, however, notably lower, whereas participant-specific model calibration yielded substantially higher performance. This indicates that while the GSR consistently reflects physiological responses linked to perceived comfort and safety, the signal exhibits a strong inter-subject variability, highlighting the necessity of personalized calibration for accurate passenger-experience assessment.