Back in 2011, we first heard about an MIT-developed system that essentially allowed robots to "see" through their fingertips. Now, the technology is being used to let them finely manipulate electrical cables.
Known as GelSight, the basic version of the system incorporates a slab of clear, synthetic rubber that is coated on one side with a layer of metallic paint. When the painted side is pressed against a surface, it deforms to the shape of that surface. Looking through the opposite, unpainted side of the rubber, one can see the minute contours of the surface, pressing up into the paint.
Using cameras and computer algorithms, the system is able to turn those contours into 3D images, that capture details less than one micrometer in depth and approximately two micrometers in width. The paint is necessary in order to standardize the optical qualities of the surface, so that the system isn't confused by multiple colors or materials.
Led by postdoctoral student Yu She, MIT researchers have now taken that technology and applied it to both finger pads of a two-fingered robotic gripper.
As a result, when that gripper closes on an electrical cable, the system is able to tell both where that cable is located on each pad, and – by monitoring friction – how quickly it's passing between them. Using that data, a computer continuously adjusts the "pose" of the gripper in three-dimensional space, and its grip strength. Adjusting the former keeps the cable centered between the pads, while adjusting the latter keeps hold of the cable while also allowing it to slide between the two pads as the gripper moves.
Among other things, the setup currently allows the gripper to grasp a loose earbud cable, slide along the length of that cable until it detects the plug at the end, adjust the orientation of that plug, and then insert it into a headphone jack. Once developed further, though, it is hoped that the technology could find uses such as inserting electrical wiring in automobiles on assemble lines, or even applying surgical sutures to patients.
The gripper can be seen in cable-manipulating action, in the video below.
Source: MIT Computer Science and Artificial Intelligence Laboratory (CSAIL)
