Tyler
Harrison Hsieh

SURF Leveraging Real-Time Eye Tracking Metrics to Infer Cognitive States During Automated Driving Mathematical/Computation Sciences

Abstract profile. Full document pending author claim.

Authors:

Tyler Harrison Hsieh

Date Created:

Not specified

Course Title:
Professor:

Not specified

About Paper:

Understanding how humans interact with automated systems such as autonomous cars is key to reduce disuse and misuse of such systems. Our research aims to study cognitive factors critical to human decision making, namely, Trust, Risk-Perception, and Mental Workload, when interacting with conditionally automated vehicles. There is little to no literature examining real-time changes in users' cognitive factors in response to different operating conditions during automated driving. Further, psychophysiological measures used to infer these states provide real-time continuous measurements, but are hard to interpret and suffer from involuntary physiological processes. We address this challenge by including eye-gaze based metrics to our measurements such as area of interest fixation identification and individual pupil diameter. Eye-tracking glasses and April Tags integrated into the driving simulation provide eye-gaze and pupillometry data that can be viewed in real time by researchers. A graphical user interface (GUI) will provide visualization through an interactive design, to clearly convey contextual information which can be used to infer some cognitive factors. This tool serves to aid researchers in validating hypotheses and providing additional insight into participant behavior in real-time. Further, unlike psychophysiological measurements, eye-gaze fixations measured in real-time suffer minimal noise due to physiological processes, and may not require inter-participant normalization. This lends itself well to autonomous vehicle usage in industry, since unobtrusive eye-tracking cameras operate in similar capacities to eye-tracking glasses and already have been implemented for basic human response detection such as drowsiness detection. Keywords: Human-Machine Interaction; Autonomous Driving; Control-Oriented Modeling; Eye Tracking

Source:

Purdue University / 2024

Topics:

No topics listed

Co-authors:

Tyler Harrison Hsieh

0