Watch This Robotic Arm With a ‘Jamming Gripper’ Shoot Hoops, Toss Darts

Jamming Gripper may sound like the name of a up-and-coming rapper but it’s really the technology that a new robot — rather a robot arm — uses to give it an effective grip that can send objects soaring at targets.

It’s nothing to scoff at. This robotic arm developed by engineers at Cornell University and University of Chicago as a DARPA-funded project can shoot hoops (accurately), toss bolts into a box, knock over little toy men by rolling a ball at them, and throw darts.

The robotic arm sends a ball rolling toward a figurine.
Robotic arm tosses darts.

Watch the machine in action:

Wired has more on how the technology works:

The robo-hand features a “universal jamming gripper,” basically a balloon filled with sand. Or, as the researchers put it, a “mass of granular material encased in an elastic membrane.” The hand can be quickly hardened by evacuating the air inside the membrane, like vacuum-packing. If the gripper is placed over a small object — a ball, a dart, a coin — the hand tightly grasps it as the air is released. By pumping air back into the hand, the robot loosens its grip and can effectively “shoot” objects a meter or so in front of it.

The point of creating such technology as the universal jamming gripper is because a robot arm will need to be able to pick up and maneuver a variety of objects, regardless of shape. The team wrote that development of a device that could perform as well as the human hand has been a challenge for engineers:

Our gripper will be useful in situations where a robot needs to grip or pick up a wide variety of items, including items it may have never seen before.  Specific applications that come to mind are: military robotics and improvised explosive device (IED) defeat missions; consumer and service robotics in unstructured environments like the home; and industrial and manufacturing robotics able to perform of a wider variety of gripping tasks than currently possible.  The gripper may provide safety benefits in the soft state it assumes between gripping tasks.

The shooting capability in particular could be useful for helping robots extend their workspace. The shooting is not accurate enough to be applicable in the high-precision tasks where many robots are employed (you couldn’t assemble a circuit board by shooting the parts from across the room for example). However, as robots move into increasingly unstructured environments (like the home), applications like sorting objects into bins or throwing away trash come to mind. Other ideas are shown in the video of our shooting demonstration.

The team plans on continuing refinement of the gripper. On a long-term scale, the team is thinking up ways to apply their knowledge of the jamming phenomenon to potentially give robots the ability to move in a different way or recover after being damaged.

[H/T SlashGear]