Photo by Chua, Yeong Jiun
Liza Dixon, M.Sc. is a doctoral candidate researching human-machine interaction in automated driving in partnership with Bosch and Universität Ulm. She coined the term autonowashing to describe the gap between the presentation of driving automation in media & marketing versus its actual technical capabilities and human supervision needs.
She has a Master of Science in Usability Engineering from Hochschule Rhein-Waal University of Applied Sciences, where she specialized in advanced driver assistance systems and trust in automation. She is an honors graduate of Flagler College where she received a B.A. in Graphic Design, and minored in Psychology and Fine Art.
Her work has been featured by The Next Web, CNBC, PAVE Campaign, The Autonocast, Guidehouse Insights, EETimes, Automotive News and others.
When she’s not thinking about future tech, you’ll find her traveling, cooking with plants, doing yoga or biking through the woods. She was born in Miami, Florida and now resides in Stuttgart, Germany.
She has a Master of Science in Usability Engineering from Hochschule Rhein-Waal University of Applied Sciences, where she specialized in advanced driver assistance systems and trust in automation. She is an honors graduate of Flagler College where she received a B.A. in Graphic Design, and minored in Psychology and Fine Art.
Her work has been featured by The Next Web, CNBC, PAVE Campaign, The Autonocast, Guidehouse Insights, EETimes, Automotive News and others.
When she’s not thinking about future tech, you’ll find her traveling, cooking with plants, doing yoga or biking through the woods. She was born in Miami, Florida and now resides in Stuttgart, Germany.
Trust in Automation:
An On-Road Study of Trust in Advanced Driver Assistance Systems
Master’s Thesis
Appropriate user trust is critical in ensuring the acceptance and safe use of Advanced Driver Assistance Systems (ADAS). Despite the prevalence of ADAS on-road today, there is a limited understanding of how trust is affected by a user’s first contact with the system on-road. Participants without prior experience were introduced to a level 2 system and completed an on-road test drive. Utilizing a mixed-methods approach including Facial Emotion Recognition (FER), trust in the system was measured at key milestones.
Experiment Proposal, Planning & Design, Moderation, Interviewing, Data Science, Statistical Analysis
Experiment Proposal, Planning & Design, Moderation, Interviewing, Data Science, Statistical Analysis
Methods
Participants were screened and selected in order to maintain homogeneousness. They completed interviews and were introduced to the vehicle systems via an introductory video, in order to ensure consistency in the information they received. Abiding by ethical and legal standards, each participant completed an on-road test drive on a pre-determined route, along with the moderator in the passenger seat.
Verbal trust scores were captured at multiple intervals. Questionnaires were administered to the participants before and after the driving experience. Multiple cameras were utilized, including roadway and driver-facing cameras for later Facial Emotion Recognition analysis.
Verbal trust scores were captured at multiple intervals. Questionnaires were administered to the participants before and after the driving experience. Multiple cameras were utilized, including roadway and driver-facing cameras for later Facial Emotion Recognition analysis.
Coding in R
Data AnalysisQualitative data, including participant behavior and commentary was recorded and aligned with factors of Trust in Automation. Questionnaire data and verbal trust scores were statistically analyzed in R. Driver-facing camera footage was analyzed by a Facial Emotion Recognition algorithm in MATLAB and further analyzed with R.
Event specific Facial Emotion Recognition (FER) Analysis
Results
Trust in Automation (TiA) scores increased in a majority of participants, and a significant shift in the factor Reliability/Competence (p=0.02) was observed post-drive. According to FER, participants with a gain in TiA and those with a loss in TiA post-drive, more frequently displayed the emotions happy and angry, respectively.
While emotional states varied greatly between drivers, patterns emerge that revealed several different types of drivers. Some showed high emotional variability while others’ emotional states remained more consistent. Driving events, for example, autonomous braking, had an effect on the driver’s emotions, giving insight into whether or not the vehicle was behaving as they expected in a given scenario.
While emotional states varied greatly between drivers, patterns emerge that revealed several different types of drivers. Some showed high emotional variability while others’ emotional states remained more consistent. Driving events, for example, autonomous braking, had an effect on the driver’s emotions, giving insight into whether or not the vehicle was behaving as they expected in a given scenario.
Feature in AutoVisionen, Herbrand Mercedes-Benz Magazine
AngelList: Self-Driving Startups
Subject Matter Expert
Autonowashing
[ aw-ton-uh-wosh-ing ]
verb. The practice of making unverified or misleading claims which misrepresent the appropriate level of human supervision required by a partially or semi-autonomous product, service or technology.
Autonowashing makes something appear to be more autonomous than it really is.
Autonowashing makes something appear to be more autonomous than it really is.
