Robots suck at picking things up. Even in industrial environments—where picking robots are fitted with various types of grippers to maneuver objects, like cans, on an assembly line—they cannot decipher between items of different shapes, sizes, and hardnesses. Unlike humans, which use visual, motor, and sensory skills to feel and pick up items, even the best robots have limited resources for the task. Cameras, actuators, and motors help, but a sensory system that could imbue robots with improved dexterity and tactile feedback would be a game-changer.
Researchers at the University of Bristol think they have a solution: 3D-printed fingertips that can help robots “feel” what they’re interacting with. The technology, described in two new papers published in the Journal of the Royal Society Interface on Wednesday, could mean the difference between machines beating humans at virtual chess, and machines being able to actually pick up and maneuver real chess pieces.