See-Thru: Towards Minimally Obstructive Eye-Controlled Wheelchair Interfaces

Corten Singer

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2018-61
May 11, 2018

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-61.pdf

Abstract—Eye-tracking interfaces are an active area of interest in human-computer interaction (HCI) research because of their potential to increase the communication bandwidth between humans and computers when the use of hands is not possible. For some people living with severe motor disabilities, the use of eyes is the only available input modality to control and interact with the various devices that enable their independence. One particularly enabling gaze-based application that has driven a large body of research is that of wheelchair navigation. The objective of this work is to develop and evaluate an eye-controlled wheelchair interface that improves upon the state of the art by considering, first and foremost, the interaction pattern between the user and the system. We explore removing the use of a computer screen from the navigation system, which normally serves to provide feedback to the user. The underlying goal behind this design decision is to avoid obstructing the user’s field of view, which is inherently limited given the nature of their disability. We present a novel eye-tracking interface device that provides feedback to the user without a screen while simultaneously allowing the user to see through it in order to provide a clear view of where they are driving. This prototype has been evaluated against the screen-based state of the art in a preliminary clinical test with three users.

Index Terms—Eye Gaze, Eye Tracking, Gaze Control, Eyes-Only Interaction, User Interfaces, Power Wheelchair, Smart Wheelchair, User Experience, Assistive Technology, Gaze Gestures, Field of View, Obstructive

Advisor: Björn Hartmann


BibTeX citation:

@mastersthesis{Singer:EECS-2018-61,
    Author = {Singer, Corten},
    Editor = {Hartmann, Björn},
    Title = {See-Thru: Towards Minimally Obstructive Eye-Controlled Wheelchair Interfaces},
    School = {EECS Department, University of California, Berkeley},
    Year = {2018},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-61.html},
    Number = {UCB/EECS-2018-61},
    Abstract = {Abstract—Eye-tracking interfaces are an active area of interest in human-computer interaction (HCI) research because of their potential to increase the communication bandwidth between humans and computers when the use of hands is not possible. For some people living with severe motor disabilities, the use of eyes is the only available input modality to control and interact with the various devices that enable their independence. One particularly enabling gaze-based application that has driven a large body of research is that of wheelchair navigation. The objective of this work is to develop and evaluate an eye-controlled wheelchair interface that improves upon the state of the art by considering, first and foremost, the interaction pattern between the user and the system. We explore removing the use of a computer screen from the navigation system, which normally serves to provide feedback to the user. The underlying goal behind this design decision is to avoid obstructing the user’s field of view, which is inherently limited given the nature of their disability. We present a novel eye-tracking interface device that provides feedback to the user without a screen while simultaneously allowing the user to see through it in order to provide a clear view of where they are driving. This prototype has been evaluated against the screen-based state of the art in a preliminary clinical test with three users.

Index Terms—Eye Gaze, Eye Tracking, Gaze Control, Eyes-Only Interaction, User Interfaces, Power Wheelchair, Smart Wheelchair, User Experience, Assistive Technology, Gaze Gestures, Field of View, Obstructive}
}

EndNote citation:

%0 Thesis
%A Singer, Corten
%E Hartmann, Björn
%T See-Thru: Towards Minimally Obstructive Eye-Controlled Wheelchair Interfaces
%I EECS Department, University of California, Berkeley
%D 2018
%8 May 11
%@ UCB/EECS-2018-61
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-61.html
%F Singer:EECS-2018-61