For a robot to be an effective tool in the workplace, it usually needs to know what sort of materials it’s working with. While it’s relatively easy to program a robot to identify only expected items, it would be helpful to push this technology further, for robots to be able to identify a variety of materials — and maybe be able to learn new ones as well.
by Keith Wagstaff
Developed by researchers at the University of Southern California’s Viterbi School of Engineering, the BioTac features a soft skin-like material set over a liquid filling, all centered around a solid “skeleton.” The faux-fingerprints help the finger pick up on vibrations as it moves across a surface, which are then detected by a hydrophone inside the “bone” and processed by a computer to figure out what the material is.
It’s actually pretty similar to how a human judges texture, except that the BioTac is even more sensitive than a human finger. The researchers had it feel 117 common materials — it correctly identified 95% of them, touching an object an average of five times. The finger can even tell which direction an object is moving and what temperature it is.
Read the full article and watch a cool video of these robotic fingers in use at Time.com. If a robot had fingers instead of grippers, hands instead of claws, what sort of new applications could they be used for? How would that change existing processes?