University of Minnesota
University Relations
http://www.umn.edu/urelate
612-624-6868
myU OneStop


Go to unit's home.

Home | DTI | 2005–06 funded proposals | Shashi Shekhar, Gordon E. Legge, Loren Terveen

Initiatives in Digital Technology: 2005–06 Funded Proposals

Shashi Shekhar, Gordon E. Legge, Loren Terveen

Indoor Navigation Aids for Visually Impaired People: Developing Cognitive and Computational Foundations: Full participation in modern society requires mobility.

Full participation in modern society requires mobility. It is so simple that most people take it for granted, but finding one's way across campus to a class, locating one's doctor's office in a large medical center, or just walking down the street to meet a friend can be difficult or impossible for people with visual impairment. Existing aids such as the white cane and guide dog are of great help, particularly in avoiding obstacles along one's way. However, they do not solve other key problems faced by visually impaired people especially for indoor navigation aids, including route planning, learning spatial layouts, and discovering landmarks or other points of interest. Thus, there is an opportunity to develop new technological navigation aids for visually impaired people.

This proposal pursues that opportunity. The overarching goal is to develop new scientific knowledge and embody it in a prototype computational navigation aid, tailored specifically for indoor navigation. Achieving this goal requires applying knowledge from a variety of specialties, solving challenging problems in each, and integrating the results; these specialties include cognitive science and computer science. Theories of spatial cognition are necessary to identify information processing subtasks (e.g., obstacle avoidance, wayfinding, and localization), environmental cues and information resources required to complete these tasks, and the cognitive demands associated with particular presentations of information. Research in human-computer interaction (HCI) is required to systematically map the space of multimodal navigation interfaces and produce particular interfaces tailored to specific users, tasks, and use situations. Closely allied research in spatial databases is required to represent spatial knowledge, query for upcoming or nearby landmarks, and compute efficient and easy-to-follow routes. Collaboration across these areas is needed to solve many of these problems, e.g., learning the structure of a new space, and to produce a working prototype navigation aid.