A universal robot gripper

Recently a colleague of mine forwarded to me a news story from the Cornell Chronicle. The topic was a robotic, mitten-like gripper being developed by physicists and engineers at Cornell and Chicago universities and iRobot, a company based in Bedford, Massachusetts (see video ).

As the video shows, the gripper can cope with an impressive variety of objects. Underlying that versatility is a surprisingly simple operating principle.

The blue latex bag that forms the gripper’s skin is filled with ground-up coffee. When the gripper first touches an object, the coffee grounds are free to flow inside the bag. Pushed gently onto the object, the gripper partially envelops it like a squishy, water-filled balloon. The grip is secured when air is sucked out of the bag, jamming the grounds so close together that they no longer flow. Frozen in an object-conforming grasp, the squishy balloon becomes a rigid claw.

When I read the release, two thoughts sprang to mind. The first concerns the physics. Coffee grounds and other granular materials are an active area of research. Unlike Bose–Einstein condensates, quark–gluon plasmas, or ocean–atmosphere interactions, granular materials are relatively easy to study in the lab.

As if to belie that convenience, however, the theory of granular materials is rich and subtle. Indeed, it remains an open question whether the jamming transition that the gripper and other systems manifest is a universal phase transition or merely a detail-dependent change.

My second thought about the gripper was biological. In its relaxed state, the gripper reminded me of an amoeba’s pseudopodia . Could higher animals have evolved grippers based on the jamming transition?

Such what-if questions are tricky to answer because a putative organ might have failed to evolve not because it is biologically infeasible but because by the time the organ could have given the animal an edge in the survival of the fittest, an organ of a different kind could have been irrevocably selected.

Is the gripper biologically infeasible? Possibly not. Fluid-filled bags abound in biology, as do molecular pumps that can move move water and other molecules in and out of cells. My hunch is that animals don’t have grippers based on the jamming transition because such grippers are biologically impractical for some reason.

Regardless of whether the universal gripper inspires you to think about science, the video is strangely compelling. A paper describing the gripper appeared yesterday in the Proceedings of the National Academy of Sciences.