In this post, I’d like to take a look at a tool that hits a little closer to home, as in, that is fairly closely related to our thesis, and follow at least partially the general idea we have.
In 2007, Mounir Bousbia-Salah and Mohamed Fezari developed a tool to help blind people navigate for the University of Annaba. Their idea was to build a simple, not-disturbing way to get from one place to the other, while all the way also checking for obstacles around the user. In order to do this, they have a computer voice tell you which ways you can go at intersections, or, as they call it “decision points”. This points the user in the right direction, and leaves him alone until the next decision point. The user can decide for him/herself where to go, and the system will keep track of this. In further research, this is supposed to happen by using GPS, but this has not been implemented as of yet. Right now the distance the person walks is measured in a rather complicated way through accelerometers, and a “footswitch” to check when the user starts a step.
The second part of their work, the one that is most resembling of our thesis, is the obstacle detection. In order to achieve this, they use ultrasound sensors, connected with vibrating elements placed on the shoulders of the user. The sensors will detect the closest obstacle, and a vibration will be generated accordingly. The have also implemented this solution in the walking cane, where obstacles are detected in the same way. This leads to an extension, albeit not a physical one, of the cane.
The idea of their solution embodies to me what an ideal solution would look like. It tries to be as little intrusive as possible, only notifying the user when absolutely necessary (for the positional navigation), or using a sense that you generally are not using while walking (for the obstacle avoidance). All the while, it does not require a huge amount of extra hardware (or at least not in the solution they want to be created eventually), which makes the adaptation for the user a bit easier. Another element that adds tot that is that they extend the usage of a very common tool for a blind person, the white cane.
Of course, there are still flaws, in the sense that there is still a decision to make about what should be “decision points”, since this alters the flexibility for the user significantly. Also, even when the GPS solution would be implemented, the device would only be useful in the Americas, since the signal on other continents is much less accurate, and might cause trouble.
As always, remarks and comments are highly encouraged, especially since this is a subject close to our own work, and any criticism could lead us to change our view on aspects from our thesis!