Robust path back-tracing guidance system for the visually impaired

Speaker Name: 
Germán Flores
Speaker Organization: 
Computer Engineering
Start Time: 
Wednesday, September 23, 2015 - 10:00
End Time: 
Wednesday, September 23, 2015 - 12:00
Location: 
280 Engineering 2
Organizer: 
Roberto Manduchi

Abstract: Indoor positioning and navigation is an active area of research due to the widespread use of devices equipped with inertial measuring units (IMUs) and wireless technologies. A system that is able to produce accurate location data may find multiple applications in location-based services (LBS) and other related fields. Location-based systems have been widely used in the healthcare, entertainment, and marketing industries to identify the location of people and objects. In the proposed research, we make use of IMUs and wireless technologies found in most smartphones to help blind people retrace the path taken inside a building and walk safely back to the initial location. This is ideal for situations in which a blind person was able to reach a certain location, for example with the assistance of a sighted guide, and needs to find his or her way back. We propose to design and develop a robust path back-tracing guidance system comprised of 3 modules: a turn detector based on hidden Markov models (HMMs) to robustly detect turns even in the presence of drift in the inertial measurements and noticeable body sway during gait; a step counter based on the state-of-the art systems that use window peak detection (WPD) or HMMs to detect the different phases during a gait cycle in order to accurately count the number of steps walked; and a Wi-Fi based positioning system to build a forward traversal map of the locations walked by a blind person in order to assist the other modules during the back-tracing guidance process. The proposed models will be trained on data collected from blind participants and designed to be usable under different phone placements on the user’s body and usage conditions. We will conduct two user studies, the first to define a baseline performance of each individual module and usability of the user interface, and the second to evaluate the overall performance of the guidance system.

 

  Center for Biomolecular Science & Engineering
1156 High St, Mail Stop CBSE,
University of California, Santa Cruz, CA 95064
831-459-1477 ext. 9-1477 | cbseweb@soe.ucsc.edu 

For questions about the UCSC Genome Browser: genome@soe.ucsc.edu

UCSC Home | BSOE Home | CBSE Home | Institute HomeInternal