Evaluation of Haptic and Audio Feedback Navigation for People with Visual Impairments
Open Access
- Author:
- Rao, Ramsha
- Graduate Program:
- Information Sciences and Technology
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 05, 2019
- Committee Members:
- Jack Carroll, Thesis Advisor/Co-Advisor
Saeed M Abdullah, Committee Member
Mary Beth Rosson, Committee Member - Keywords:
- Remote Sighted Assistance
Assistive technology
Haptic feedback
Visual impairment - Abstract:
- Assistive technologies are becoming prevalent among people with visual impairment (PVIs); by definition, such supports can increase these individuals’ autonomy in their daily lives. Examples include Artificial Intelligence powered applications, applications using Remote Sighted Assistants, Braille Technology, Screen Magnifiers, and Readers among many others. Among these technologies, Remote Sighted Assistant applications are garnering considerable attention; these use the novel paradigm of a conversational agent who guides the PVIs through a video-mediated conversation. My thesis project investigated novel augmentations for a Remote Sighted Assistance service called AIRA. The specific goal was to support PVIs in navigation within the task context of grocery shopping (moving around a store but also acquiring target products). Working with a larger team, I investigated a wristband that provides haptic (vibration) feedback (with the help of the four attached vibration motors that are controlled by an Android application) to the participant for obstacle warning and hand navigation. The haptic information is combined with audio (spoken) feedback from the Remote Sighted Assistant who communicates through a video call; this enables the agent to see what is in front or around the participant and provide corresponding guidance. The prototype was tested on sighted, blindfolded participants to examine its strengths and weaknesses and to gather suggestions for future device improvements. The valuable feedback received from the participants will be helpful for both our own future work and that of other researchers working on these kinds of systems.