I had the honor of knowing Bill Dobelle for more than 25 years.  With his ever-present bowtie, he was enamored with bringing sight to the blind (which number about 15 million people around the world).  I am sure if he hadn’t died prematurely at 64, he would have succeeded completely.   As it was, he managed to enable those who had lost their sight to recognize shapes, patterns, and colors, using a series of implanted electrodes.

Now, some folks are using the technology that has been developed for robots to enable the blind to navigate.  Dr. E. Pisaloux (Pierre and Marie Curie University, Paris) is leading a team that has developed one such device. Presenting the results at MIT, he described his system: The wearer of the specialized glasses (with cameras on each side of the glasses, as well as sensors) is presented with a three-dimensional map of the local environment.  This spatial  information included the depiction of the exact location of the eyeglass wearer (the subject) within that environment.  The images are processed, denoting walls, objects, and the like, as well as the speed of the subject (via accelerometers and gyroscopes). Processing some 10 maps per second for the subject,  the device transmits this information to a Braille device (an 8 cm square grid) that affords the blind the ability to navigate on their own.

While not a repurposed robot technology, the efforts of James Loudin and Daniel Palanker (Stanford Unriversity) and nine other colleagues are updating Dobelle’s concept (above), as reported in the journal, Nature Photon.  Instead of wired connections to electrodes in the brain, they employ photovoltaic pixels and wireless connections.  Because these “solar panel” pixels are only 70 um (1/3 the thickness of a strand of hair), they can respond to one photon of infrared light, and thereby light up individually, just like our eyes’ photoreceptors operate.  The testing included implanted photovoltaic pixels into the retinas of rats (whose photoreceptors had been destroyed for this test), and verified that the light was converted into electrical energy when light impinged upon the receptors.  That worked- but the system, right now, provides images that are closer to what our eyes see when we rub them- starbursts of colors and imagery at random.  But, it is a first step in this approach.

Drs. Nanayakkara. Shilkrot, and Maes of MIT’s Media Lab are using a yet different approach developing what they call an EyeRing.  By pointing the finger-worn device at an object, the subject can receive information- via audio- about the target.  Using voice commands, the subject lets the device know what information is desired.  The ring, which is worn on a finger (right now, it’s a pretty big ring, about 3 cm tall and 1.5 cm on the sides, but still less obtrusive than a cane) takes a picture, which is transmitted to a smartphone, which analyzes the image and provides the audio information via an earpiece to the subject.   One of the coolest things it can do right now is let the blind person know what bills s/he is taking out of his wallet or the colors of various pieces of clothing.

Stay tuned for the next series of  improvements.  The world needs Dobelle’s vision to come to light!

Share this: