Tuesday, September 4, 2012

Paper Reading #2: The User as a Sensor - Navigating Users with Visual Impairments in Indoor Spaces using Tactile Landmarks


Intro -

Title:
     The User as a Sensor: Navigating Users with Visual Impairments
                     in Indoor Spaces using Tactile Landmarks

Reference Information:
     CHI 2012, May5-10, 2012, Austin, Texas, USA

Author Bios:
     Ilias Apostolopoulos and Navid Fallah have been working in the PRACSYS (Physics-aware Research for Autonomous Computational SYStems) Group of the Robotics Research Lab at the University of Nevada, Reno, towards their PhD's for a year.  They study under the guidance and direction of Associate Professors Kostas Bekris and Eelke Folmer who share the title of Director at the UNR Robotics Research Lab.

Summary -

This paper evaluates a system called Navatar that visually impaired users can use to help them locate and navigate around an indoor environment using tactile landmarks.  Previous indoor systems required expensive alterations to the environment or sensing and computing equipment which has prevented large scale implementation.  Navatar uses the accelerometers in a user's smartphone coupled with an annotated virtual representation of the indoor environment in order to guide the visually impaired user through a sequence of tactile landmarks.


"A user study with six visually impaired users evaluated the accuracy of Navatar and found that users could successfully complete 85% o."

Related Work -
  1. Mobility in Individuals with Moderate Visual Impairments -  http://psycnet.apa.org/psycinfo/1990-23089-001
  2. Personal Guidance System for People with Visual Impairment: A Comparison of Spatial Displays for Route Guidance -  http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2801896/
  3. Drishti: An Integrated Navigation System for Visually Impaired and Disabled -  http://www.harris.cise.ufl.edu/projects/publications/wearableConf.pdf
  4. Indoor Wayfinding: Developing a Functional Interface for Individuals with Cognitive Impairments -  http://informahealthcare.com/doi/abs/10.1080/17483100701500173
  5. An Integrated Wireless Indoor Navigation System for Visually Impaired - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5929098
  6. RFID in Robot-assisted Indoor Navigation for the Visually Impaired -  http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1389688
  7. Comparing Methods for Introducing Blind and Visually Impaired People to Unfamiliar Urban Environments -  http://cogprints.org/1509/
  8. The Development of the Navigation System for Visually Impaired Persons -  http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1020485
  9. A Model-Based, Open Architecture for Mobile, Spatially Aware Applications -  http://dl.acm.org/citation.cfm?id=719105&CFID=109411254&CFTOKEN=12089765
  10. Where did that Sound come from? Comparing the Ability to Localise Using Audification and Audition -  http://informahealthcare.com/doi/abs/10.3109/17483107.2011.602172
The previous works I found that were similar in topic, as well as the ones listed in the paper's references, all show that the specific way they are attempting to navigate a blind user through an annotated virtual representation of the users local indoor area using accelerometers and other sensors is truly 'novel'.  They presented one previous paper as a case study to determine if this method could work for visually impaired (blindfolded) users, but now are presenting a new paper that has shown effectiveness for actual blind users.

Evaluation -

This second paper on Navatar presents studies conducted to determine the effectiveness of this system used for indoor navigational purposes by blind users.  They conducted the study on 6 participants and received both quantitative and qualitative feedback.  There were 11 paths tested, and the system showed to help successfully navigate the user through the physical landmarks 85% of the time.  Also, the users were able to answer open-ended questions after the study to collect qualitative data and assessed a 5-point Likert scale that showed an average liking of 4.66 out of 5 for each category.

Discussion -

Although Navatar did make navigation more efficient and reduce the overall effort by the user, there were several areas that were noted could be sufficiently improved.  Since an annotated virtual representation of the indoor space is needed to be able to mark tactile landmarks, robotic mapping could be used to gather this information in real-time to aide a user in an unfamiliar setting.  Also, another improvement was simply using a headset in order to keep the hands free while blindly navigating the terrain.


"The application provides directions through text to speech using the smartphone's speaker and the user confirms executing each direction by tapping the screen.