Bringing science and development together through news and analysis

  • Mexican team builds navigation device for blind people

Image credit: Jan Banning/Panos

Speed read

  • Data from pairs of cameras and ultrasonic sensors are sent to a tablet computer

  • This builds a three-dimensional image and then provides audible directions

  • The device also uses artificial intelligence to recognise objects and colours

[BUENOS AIRES] Mexican researchers have developed an artificial intelligence navigation system for partially or totally blind people that can sense its surroundings in three dimensions.

The prototype device is designed to allow users to freely move from one point to another avoiding both static and moving obstacles, and can learn to recognise colours and utility bills.

It looks like an adapted pair of glasses and has two sensors that locate objects around the user by bouncing ultrasonic waves off them. It also has two cameras and its batteries allow continuous use for about four hours.

“This is a very attractive product for users with limited vision because different sensors have been fused allowing the avoidance of obstacles.”

Taihú Pire, Buenos Aires University


The device is also equipped with a GPS (global positioning system) to help users navigate when outdoors.

Everything plugs into a tablet computer that processes image and sound data and uses it to calculate distances and build up a three-dimensional image of the user’s surroundings. The tablet acts in a similar way to a car sat nav, providing navigation instructions and warnings through earphones.

The tablet uses artificial intelligence software to recognise navigation routes and signs.

Its lead developer is Eduardo Bayro, a researcher at the Center for Research and Advanced Studies in Mexico. He is already well known in the country, having developed a humanoid robot, called Mexone, in 2011.

The prototype took five years to develop. During this period, it was successfully tested by a group of blind girls in the Mexican city of Guadalajara. Bayro tells SciDev.Net that the girls were asked to navigate between different indoor locations and recognise specific credit cards, utility bills and bank notes.

Mexico’s National Council for Science and Technology gave the project a total of US$1.7 million of public money between 2010 and 2013.

Bayro says the device will be on the market early next year and will be commercialised by the a company called Qualtop.
 
Similar devices already exist, including low-tech prototypes developed by inventors in Africa.

But Bayro says the Mexican device is unique in several ways. For example, its use of ultrasound to detect translucent obstacles such as glass and its use of artificial intelligence to recognise objects, signs and places.

The fact that the device will be turned into a commercial product also makes it important, says Taihú Pire, a researcher in the Laboratory of Robotics and Embedded Systems at the Buenos Aires University in Argentina.

“This is a very attractive product for users with limited vision because different sensors have been fused allowing the avoidance of obstacles and [there is] the GPS for location,” Pire says.

The device may cost about US$1,500 once on sale, says Bayro, but he thinks it could be made US$300 more cheaply than that after further developments.

However, he recognises that without support from public institutions, the device would remain largely unaffordable. He says he is currently “looking for social programmes that can finance access to the lenses in Bolivia, Cuba and Mexico”. 
Republish
We encourage you to republish this article online and in print, it’s free under our creative commons attribution license, but please follow some simple guidelines:
  1. You have to credit our authors.
  2. You have to credit SciDev.Net — where possible include our logo with a link back to the original article.
  3. You can simply run the first few lines of the article and then add: “Read the full article on SciDev.Net” containing a link back to the original article.
  4. If you want to also take images published in this story you will need to confirm with the original source if you're licensed to use them.
  5. The easiest way to get the article on your site is to embed the code below.
For more information view our media page and republishing guidelines.