USING VIRTUAL SPATIALIZED OUTPUT TO AUGMENT AND IMPROVE THE USER INTERFACE OF GPS NAVIGATION SYSTEMS FOR PEOPLE WHO ARE BLIND

J. R. Marston¹, J. M. Loomis², R. G. Golledge¹, R. L. Klatzky²

1 - Department of Geography, University of California

2 - Department of Psychology, University of California

j.rowell@apu.ac.uk

Currently there are a number of research projects on providing effective GPS navigation systems that can be used by people who are blind or visually impaired. A few commercial systems are already available. Most of these projects devote little attention to the question of how best to display spatial information to the blind traveler.

Since 1985, a team at the University of California at Santa Barbara, led by Jack Loomis, has been developing and evaluating various ways of providing spatial information. Most systems today do not make use of electronic compasses and thus cannot know the user’s facing direction, except when the user is walking straight. Using a compass with specialized hardware, our system allows for several different types of “spatialized displays” whereby the user can obtain direct perceptual information, other than that provided by speech, about waypoints along the route and important on-route and off-route locations. Over the past 6 years, we have conducted different formal experiments investigating different spatialized displays. I will describe some of this research on path following in open areas, on city sidewalks, and through a park. The displays to be discussed include a virtual sound display that allows users to hear sound seeming to come from locations in the environment, a hand-help Haptic Pointing Interface, a body mounted interface, and a hand-held receiver that can give both GPS information and also function as a Talking Signs^(R) Remote Infrared Audible Signage pointing device that can be used indoors, and other variations.

Current GPS for the blind offer little or no choice for user options. I will report on experiments where users were able to control the amount of spatial information they needed in different types of environments; such as a bounded city sidewalk grid, a semi-bounded walk along paths in a park, and in totally unbounded open areas. This information, along with user ratings about the hardware and their desire for different options provide robust data for the design of future Personal Navigation System for the blind and visually impaired.

Without vision, many problems arise when trying to replace maps, as most spatial information must be delivered through some type of spatial language. Successfully using spatial language requires cognitive processing, memory, and a need to have good spatial skills to understand and encode the relevant information. Our research on “spatialized displays” has shown that map information can be successfully transferred to a blind person.