Date of Award

Fall 2023

Project Type

Dissertation

Program or Major

Electrical and Computer Engineering

Degree Name

Doctor of Philosophy

First Advisor

Andrew Kun

Second Advisor

John Lacourse

Third Advisor

Md Shaad Mahmud

Abstract

Future automated vehicles will allow drivers to reclaim some of their driving time and perform personal or work-related activities while the car is in automated driving mode. However, traditional automotive user interfaces (UIs) are not designed to support such activities. For a vehicle to be considered a safe and productive workspace, we will have to explore how drivers can interact with emerging UIs in a car to engage in complex non-driving related tasks (NDRTs) and safely resume driving when needed. In the second chapter of this thesis, we present a user-elicitation study where we investigate how drivers would want to use gestures and voice commands to interact with augmented reality windshield displays in highly automated vehicles. We argue that it is important to evaluate interaction modalities from the users’ point of view before designing unconventional UIs for future automated vehicles. In chapter three, we examine what strategies people use while switching from NDRT to driving. We identified two common takeover strategies (suspension and interleaving) and show that it is important to examine takeover strategies in addition to takeover performance to fully understand takeover in automated vehicles. In the fourth chapter, we present findings from two driving simulator studies. In these studies, we analyze how different factors influence what strategies drivers use during takeovers and the relationship between these strategies and takeover performance. We found that people are more likely to interleave between driving and NDRT when taking over if they are asked to prioritize NDRT or allowed a longer time to take over. We also found that the effect of priority is moderated by the takeover time budget. We did not find any relationship between the takeover strategy and takeover quality in terms of lateral and longitudinal vehicle control while driving in a simple traffic scenario. Drivers took longer to take over driving but glanced at the driving scene faster while following the interleaving strategy compared to the suspension strategy.

Download

Share

COinS