According to the Braille Institute of America [1], there are 15 million blind and visually impaired people in the United States. They have different important daily needs including navigation. Visually impaired people rely on different tools and skills to navigate. They usually rely on their white canes, seeing-eye-dogs and other skills acquired over time to aid their navigation. Many technologies have been developed to assist blind users with different navigation needs. These include obstacle avoidance technologies as well as routing technologies. Any routing technology for blind requires different components including localization, map representation, path planning, interface, and a component to translate the planner output into meaningful instructions. The focus of this work is on developing the interface and the translator component of a full smart phone-based navigation system, called NavPal. The application improves on previous work by giving better quality instructions, and giving more flexibility to the user in choosing the level of verbosity and using different input/output modalities. The application tries to keep a good balance between the quality of the navigation instructions and the automatic production of these instructions. The interface was tested with eight blind users who traversed three routes, each. The results indicated that 75% of the twenty four navigation tasks were accomplished successfully, while relying only on the interface instructions. The users provided feedback on all components of the interface and provided suggestions for improvement, which will be considered in future work.

81 pages

Return to: SCS Technical Report Collection
School of Computer Science

This page maintained by reports@cs.cmu.edu