Harvard Develops Robots for Grasping Fragile Objects

As robots become more specialized, the need for new and innovative expressions is more apparent than ever. Harvard researchers have now developed a robot with a soft, tentacle-based grip designed to maneuver delicate objects safely. It greatly extends the range of what objects can be safely gripped.

Simplifying fragile grip technique for robots

Right now, robotics leverages a balance of operator expertise, embedded sensors, and feedback loops. Others use machine learning algorithms to adjust grip strength and technique. Researchers at Harvard’s John A Paulson School of Engineering and Applied Sciences have taken a completely different approach.

The new robotic gripper takes inspiration from nature. Similar to the way jellyfish entangle prey, the robotic arm entangles and grabs objects with a network of thin filaments. The filaments don’t use any sort of sensing or feedback control. Instead, they use simple inflation to wrap around the target. One side of the tube features thicker rubber than the other side. When inflated, the tubes form curls that entangle the object.

See also: Robots Aiding Digital Transformation in Textiles

A safer approach based on nature

The filaments aren’t very strong alone, but as they curl, they increase the entanglement of the hold and, thus, its strength. Even though the hold is secure, individual filaments aren’t strong enough to damage the target.

This is an entirely new approach based on previous research. It replaces complex actions with simple physics, making it possible to create strong, secure holds using a streamlined action. This could expand robotics use cases far beyond the typical and into previously impossible actions.

Some possible real-world applications include handling fruit, delicate tissue in medical procedures and settings, and even fragile or highly irregular objects such as a wine glass. Depressurizing the filaments releases the hold safely.

Researchers used simulations and controlled experiments to test efficacy with positive results. We soon could see robots maneuvering objects we never imagined.