BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251016T135144Z
LOCATION:Grand Hall J
DTSTART;TZID=America/Chicago:20251014T141000
DTEND;TZID=America/Chicago:20251014T143000
UID:HFESAM_ASPIRE 2025_sess261_LECT576@linklings.com
SUMMARY:Measuring Trust Degradation in AVs: Opportunities and Challenges U
 sing Body Posture and Deep Learning
DESCRIPTION:Cherin Lim, Tianhao Xu, and Prashanth Rajivan (University of W
 ashington)\n\nTrust is a critical factor in the adoption of autonomous veh
 icles (AVs), yet critical errors (e.g., cybersecurity incidents) to AV ope
 rations can severely undermine it. Traditional trust assessments rely on s
 ubjective questionnaires, which may lack granularity and real-time applica
 bility. This study proposes a novel approach to trust measurement by devel
 oping deep learning models using drivers’ body posture data. A driving sim
 ulator experiment with 40 participants was conducted across three drives: 
 baseline, error drive (errors from simulated cybersecurity threats), and p
 ost-attack. Participants’ body posture data were captured via video and pr
 ocessed to extract the x-y coordinates of key body points. Subjective trus
 t levels were collected from questionnaires at the end of each drive, and 
 were used to train and evaluate Long Short-Term Memory (LSTM) and Convolut
 ional Neural Network (CNN) models. The LSTM model outperformed the CNN, ac
 hieving up to 96% accuracy by effectively capturing temporal patterns in p
 osture linked to trust changes. Results suggest body posture is a viable, 
 real-time indicator of trust, particularly when modeled with temporally se
 nsitive architectures like LSTM. This approach offers a direction for adap
 tive AV systems capable of responding to dynamic trust fluctuations.\n\nTr
 ack: Surface Transportation\n\nSession Chairs: Shannon Roberts (University
  of Massachusetts, Amherst) and Stanislaus Oshimeje (Michigan Technologica
 l University)\n\n
END:VEVENT
END:VCALENDAR