Sep 30, 2015 | By Kira

When you picture robotic hands, the image that most likely comes to mind is a cold, hard metallic claw that clamps onto its target and crushes the life right out of it—regardless of weather the object is a titanium rod or a helpless teddy bear. Now, a new generation of modular, 3D printed soft robot grippers can easily pick up over 70 everyday times of different weights and sizes, ranging of stuffed animals to eggs to a single sheet of paper, without damaging or dropping them. The robot was designed by MIT’s Computer Science and Artificial Intelligence Lab (CSAIL), and could eventually allow robots to not only pick up, but actually use a variety of tools designed for human use.

The soft and highly compliant hand consists of three individual 3D printed fingers that can be attached to existing robots. Each finger is outfitted with resistive bend sensors that collect data points, just one from each finger. These points are configured using a clustering algorithm, allowing them gripper to identify the object in front of it and apply the requisite amount of force to grip and lift it off of a surface. According to the researchers, this breakthrough technology is the first step towards robust prorprioceptive soft grasping.

“If we want robots in human-centered environments, they need to be more adaptive and able to interact with objects whose shape and placement are not precisely known," said Daniela Rus, director of CSAIL. “Our dream is to develop a robot that, like a human, can approach an unknown object, big or small, determine its approximate shape and size, and figure out how to interface with it in one seamless motion.”

The fabrication of each finger is based on a lost-wax casting process, which beings with 3D printing a set of model and mold parts. The developers also prioritized a modular interface in order to enable the gripper to be usable for a variety of existing hardware bases, simply by swapping out the 3D printed interface.

Attaching a soft finger onto the 3D printed interface

As the video below shows, the impressive gripper can truly pick up just about anything. Eggs, tennis balls, Rubik’s cubes, water bottles, paper cups, stuffed animals and hair accessories are no match at all. It can even pick up a CD and piece of paper lying flat on a table, and, if the object is upright, it can lift such thin and delicate items as a pen, glasses or a long Q-tip. In each of the tests, the robot had no prior knowledge of which item was being picked up; it was simply commanded to close its fingers all the way, and thus had to rely on the clustering algorithm to choose the appropriate strength and grasping method (enveloping grasp for larger items, pinching grasp for lightweight ones).

The researchers compared their robust soft gripper to a default rigid gripper. In their tests, the default gripper was unable to pick up the CD or piece of paper, and crushed the paper cup and soda can to oblivion. The soft gripper, on the other hand, performed each task without issue. There are still some kinks to work out—for example, it cannot properly grip items that are slippery or heavy, such as small books—but with future research, that might not be a problem for long.

“Future work will take these core principles and methods and expand them to create a more robust and capable gripper,” said the CSAIL team. They plan to add resolution with better flex sensors as well as multiple internal flex sensors to get independent data from different segments of each finger.

With this technology, we may soon be able to use robots for a much wider variety of tasks—from picking up soft items, such as our laundry; delicate items, like our dishes; or performing even more precise and intricate tasks, like maneuvering medical instruments. “This knowledge is useful for creating a system which can use objects in more complex ways,” said the researchers. “Rather than just performing pick and place operations, robots should be able to pick up a variety of tools designed for human use and be able to handle them appropriately.”

The research was published in a paper titled "Haptic Identification of Objects using a Modular Soft Robotic Gripper" by Bianca S. Homberg, Robert K. Katzschmann, Mehmet R. Dogar, and Daniela Rus, and presented at the International Conference on Intelligent Robots and Systems in Germany.

 

 

Posted in 3D Printing Applications

 

 

 

 

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