A robot gripper hand, invented by engineers at Cornell University, has inspired some of the same researchers to write an algorithm that can teach any robot how to pick up oddly-shaped objects.
The algorithm allows a robot to learn complex grasping skills by trial and error. The robot can then apply what it has learned in situations it hasn't previously encountered. The researchers said the method is hardware agnostic and will work with any type of robot gripper. This could be especially useful for autonomous industrial robots used in assembly or palletizing lines.
An algorithm inspired by Cornell University's universal jamming gripper robot hand, shown here picking up a glass of water, can teach any industrial robot how to pick up unfamiliar, oddly-shaped objects. (Source: John Amend, Cornell University)
The original "universal jamming gripper" hand was developed in October 2010 by the Cornell Creative Machines Lab headed by Hod Lipson, associate professor of mechanical engineering and computer science. It consists of a large latex balloon filled with a granular material, such as ground coffee. By modulating the air pressure inside the balloon, the granular material can quickly harden or soften to adapt to the object the arm is attempting to grasp.
The new algorithm was developed by Lipson and Ashutosh Saxena, Cornell assistant professor of computer science and a specialist in machine learning. Using the algorithm, a robot employs a 3D image of an object to examine several rectangles that match the size of the gripper. The robot tests each rectangle on a variety of features, and is also trained with images of different objects.
During this process, the robot builds up a library of features that are common to the properties of "good-grasping" rectangles. When the robot is given a new object, it chooses the rectangle that has scored the highest, based on the rules it has developed for good grasping. The robot also considers the overall size and shape of the object to help it choose a stable grasping point.
Agree....Most of the comments are based on environments where uniform parts are pre-aligned. Many times that's fine, but what if electronic components, gears, etc. could be "loose" and gripped and oriented by more sophisticated robotics? It could result in net savings. Another application is when the component shapes or orientation are irregular and poorly defined- logs, chicken wings, gemstones, or debris on the seabed.
Jack, that's a good point about the use case of slight changes in the expected location of the object to be picked up. The main advantage the researchers cited was in adapting to different shaped and oddly shaped objects and being able to pick them up without dropping them (or spilling water from them as shown in the photo).
Yes, Ann, the adaptation to different shapes is the key component of the algorithm. I see two practical applications for something like that. First, it gives the robot a much higher margin of error when moving a product. If the product is not quite in the right orientation or has moved somewhat from where it it expected, the gripper can still get it (within reason). The second application is if the product the robot is trying to grab gets redesigned. A minor modification to it physical shape may not require as drastic of changes to the processes if the robot is still able to adapt to it.
jmiller, you can see and hear about the inner workings of the gripper's ball--what makes it a jamming gripper--in the video linked to in the article. The fact that the robot has to follow the same repetitive steps is secondary here: it's the fact that it may have to adjust those repetitive steps to different shaped objects, as stated in the article. That's what the algorithm teaches it to adapt to.
This gripper--which is not the main subject of the DN article--is not designed to pick and place small chips or other tiny objects on a high-speed line. The universal jamming gripper is a very different gripper designed to quickly grasp and release, or throw, a wide variety of object shapes. According to a FAQ http://creativemachines.cornell.edu/jamming_faq_2 for an earlier IEEE article about this gripper by its inventors, not the algorithm which my article focused on, specific applications include "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." According to that article, universal grippers can be used for sorting and throwing objects. One immediate use that comes to mind is end-of-line palletizing for non-fragile objects. A different (non-jamming) approach to universal grippers is shown here: http://blog.robotiq.com/bid/29474/Universal-Gripper-Tooling-for-Pre-Engineered-Robotic-Cells
Are they robots or androids? We're not exactly sure. Each talking, gesturing Geminoid looks exactly like a real individual, starting with their creator, professor Hiroshi Ishiguro of Osaka University in Japan.
Truchard will be presented the award at the 2014 Golden Mousetrap Awards ceremony during the co-located events Pacific Design & Manufacturing, MD&M West, WestPack, PLASTEC West, Electronics West, ATX West, and AeroCon.
For industrial control applications, or even a simple assembly line, that machine can go almost 24/7 without a break. But what happens when the task is a little more complex? That’s where the “smart” machine would come in. The smart machine is one that has some simple (or complex in some cases) processing capability to be able to adapt to changing conditions. Such machines are suited for a host of applications, including automotive, aerospace, defense, medical, computers and electronics, telecommunications, consumer goods, and so on. This discussion will examine what’s possible with smart machines, and what tradeoffs need to be made to implement such a solution.