Mehta, Utkal V. and Alim, Mohammed Zainal and Kumar, Shivneel (2017) Smart Path Guidance Mobile Aid for Visually Disabled Persons. Procedia Computer Science, 105 . pp. 52-56. ISSN 1877-0509
Preview |
PDF
- Published Version
Download (476kB) | Preview |
Abstract
A traditional blind-navigation cane mostly used by a visually impaired person is not very appropriate mainly due to narrow search area. While a
conventional cane warns of changes along the ground, it does not warn of other walking hazards and objects above a person’s waist. There are
many electronics based blind-navigation devices employ a voice guided GPS (global positioning system) and/or complex high-order processor.
It is apparent that the costs of these devices are too high that a common visually impaired people cannot afford them. In addition, a kind of
previous arts is difficult to handle due to the weight, volume and functions incubated with basic purpose. Therefore, these types of advanced
navigation systems are difficult to be commercialized. The purpose of this research is to design and develop a smart path guidance system for the blind and visually impaired, particularly the mobile aid to carry by hand, contains a smart sensor logic system. An appropriate model has developed for the selected design with embedding fuzzy logic decision. A presented solution is also tested for various condition inputs to verify
the system’s behavior. Through several experiments, the sensors are calibrated to increase the accuracy of decision. The presented prototype
enables the blind person to walk freely in an unfamiliar environment.
| Item Type: | Journal Article |
|---|---|
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > Robotics and Automation |
| Divisions: | Faculty of Science, Technology and Environment (FSTE) > School of Engineering and Physics |
| Depositing User: | Utkal Mehta |
| Date Deposited: | 13 Mar 2017 01:53 |
| Last Modified: | 13 Mar 2017 01:53 |
| URI: | https://repository.usp.ac.fj/id/eprint/9632 |
Actions (login required)
 |
View Item |