Oct 11, 2017 | By Benedict

Engineers at the University of California San Diego have developed a partially 3D printed soft robotic gripper capable of 3D scanning the object it is gripping. 3D printing was used to make the gripper’s actuators.

Sometimes when you’re scrambling around in the dark for a light switch, a pair of glasses, or—let’s go out on a limb here—a 3D printer, you might wish that your sense of touch was a little better. Overall, however, we as humans are pretty good at knowing what’s in our hands—even when we can’t see anything.

Things are a bit different if you’re a robot, and that can make some of your general robotic tasks a bit tricky. If you’re a robotic gripper, for instance, you might have to pick up a whole range of different objects, with different shapes, textures, and strengths.

Clearly, the nature of the gripped object matters a lot in the process of gripping, because different kinds of grip will be needed for a stone and, say, a delicate glass ornament.

So how do you get a gripper gripping the right way? Of course, humans can simply do the hard work, programming robots to grip harder or softer for particular tasks. But that’s not especially useful for automated tasks—a robot that needs constant supervision is less valuable than one that can work independently.

In an ideal world, a robot responsible for gripping different things should be able to know exactly what it’s gripping, and exactly how to grip it.

An engineering team at the University of California San Diego has designed and built a gripper that can pick up and manipulate objects without either needing to see them or needing to be trained.

Discerning objects without “seeing” them is particularly useful, these engineers say, because sometimes low light conditions mean that a camera would not be of use. Instead, this gripper uses a kind of 3D scanning technique to determine what’s in its grasp.

Led by Michael T. Tolley, a roboticist at the Jacobs School of Engineering at UC San Diego, the engineering team presented the scanning gripper at the International Conference on Intelligent Robots and Systems (or IROS) at the end of September in Vancouver, Canada.

At the conference, the clever gripper was added to an industrial Fetch Robotics robot, which was then able to pick up a range of objects, including both delicate and rugged things, with the appropriate grip.

The gripper has three fingers, each of which is made of three soft flexible pneumatic chambers that move when air pressure is applied. These soft chambers allow the gripper to twist, screw, and perform other dextrous motions that would be impossible with a hard robotic claw. The gripper can also operate with several degrees of freedom.

"We designed the device to mimic what happens when you reach into your pocket and feel for your keys," said Tolley, but the process is also a bit like a CT scan, using multiple 2D “slices” to form a complete 3D image.

A silicone “skin” on each finger acts as a kind of sensing aid, with embedded conducting carbon nanotubes changing conductivity as the fingers flex. This lets the silicone skin detect when the fingers are moving and coming into contact with an object, which ultimately enables a computer to build up a picture of the shape of an object.

3D printing was used to create actuators for the gripper’s fingers. “Using a photopolymer resin (Veroclear, Stratasys Objet350 Connex3), we 3D printed actuator molds which consisted of five pieces that encapsulate and shape the silicone during the curing process,” the researchers explained.

They added that 3D printing on the high-end Objet350 Connex3 produced far better results than printing on an FDM printer.

The engineering team plans to keep working on the robotic gripper, and could even incorporate more 3D printed finger elements in future models in order to make the system more durable. The real test, however, will involve turning the gripper into something that can recognize and identify objects, rather than just model their 3D shape.

This ability will ultimately be made possible using machine learning and artificial intelligence.

The research team’s study, “Custom Soft Robotic Gripper Sensor Skins for Haptic Object Visualization,” can be read here.



Posted in 3D Printing Application



Maybe you also like:


Leave a comment:

Your Name:


Subscribe us to

3ders.org Feeds 3ders.org twitter 3ders.org facebook   

About 3Ders.org

3Ders.org provides the latest news about 3D printing technology and 3D printers. We are now seven years old and have around 1.5 million unique visitors per month.

News Archive