A UMan robotic arm
Researchers at the University of Massachusetts Amherst have developed a mobile robotic arm which can manipulate objects by 'seeing' its environment through a digital camera. This robotic arm, dubbed UMan, or UMass Mobile Manipulator, can 'approach unfamiliar objects, such as scissors, garden shears and jointed wooden toys -- and learn how they work by pushing on them and observing how they change' -- like a child discovering the world. As said the leading researcher, 'Mobile robots play an important role in many settings, including planetary exploration and manufacturing. Giving them the ability to manipulate objects will extend their use in medical care and household assistance.' But read more...
You can see above the "UMan learning about the kinematics of a tool by interaction." (Credit: UMass Amherst Robotics and Biology Lab)
videos
And you can see above the "UMan applying its knowledge about the kinematics of the tool to perform a task (create a right angle between the tool's links)." (Credit: UMass Amherst Robotics and Biology Lab) These two images have been extracted from short videos available on this page.
This research project has been led at UMass Amherst Robotics and Biology Lab by Dubi (Dov) Katz, a doctoral student of computer science, and Oliver Brock, a professor of computer science. But the whole robotics group was involved in this project.
According to Katz, "UMan sits on a base with four wheels that allow it to move in any direction, and a system of lasers keeps it from bumping into objects by judging their distance from the base." He says that it was relatively easy to teach it how to walk.
But it was much harder to teach it how to manipulate objects. "'Robots in factories perform complex tasks with ease, but one screw out of place can shut down the entire assembly line,' says Katz, who recently met with representatives from Toyota Motors. 'Giving robots the same skills as humans turned out to be much more difficult than we imagined, which is why we don’t have robots working in unstructured environments like homes.' The key was giving the UMan eyes in the form of a digital camera that sits on the wrist. Once they added the camera, which coupled manipulating objects with the ability to 'see,' the complex computer algorithms needed to instruct the UMan to perform specific tasks became much simpler."
For more information about this research project, please visit this page about Autonomous Mobile Manipulation (AMM). "Autonomous Mobile Manipulation could be considered the holy grail of robotics: a successful approach to autonomous mobile manipulation has to integrate virtually every aspect of robotics, including mechanism and sensor design, low-level control, motion planning, machine learning, computer vision, and reasoning. Currently, most of the available robotic applications are either capable of performing a very limited set of skills, or can only operate in relatively structured environment. The task of bringing robots to houses and to our daily life remains a challenge. The commercial potential is promising. Truly autonomous robots could dramatically affect health care and planetary exploration. Moreover, those robots could contribute to logistic and military applications, maintain satellites in orbit, and many other tasks currently performed by humans. Research in AMM is also important in terms of advancing basic science."
But if you really want to learn more about this research work, here is a selection of technical papers produced by the team.
- How Can Robots Succeed in Unstructured Environments? presented at a workshop on Robot Manipulation: Intelligence in Human Environments at Robotic: Science and Systems, Zurich, Switzerland, June 2008 (PDF format, 6 pages, 190 KB)
- Learning to Manipulate Articulated Objects in Unstructured Environments Using a Grounded Relational Representation in the Proceedings of Robotics: Science and Systems, Zurich, Switzerland, June 2008 (PDF format, 8 pages, 3.36 MB)
- Manipulating Articulated Objects With Interactive Perception, in the Proceedings of the IEEE International Conference on Robotics and Automation, Pasadena, USA, May 2008, Best Manipulation paper award finalist (PDF format, 6 pages, 1.21 MB)
- The UMass Mobile Manipulator UMan: An Experimental Platform for Autonomous Mobile Manipulation presented at a wWrkshop on Manipulation for Human Environments at Robotics, Science and Systems, Philadelphia, Pennsylvania, USA, August 2006(PDF format, 8 pages, 1.74 MB)
Here is the abstract of the 3rd paper. "Robust robotic manipulation and perception remains a difficult challenge, in particular in unstructured environments. To address this challenge, we propose to couple manipulation and perception. The robot observes its own deliberate interactions with the world. These interactions reveal sensory information that would otherwise remain hidden and also facilitate the interpretation of perceptual data. To demonstrate the effectiveness of interactive perception we present a skill for the manipulation of an articulated object. Using this skill, we show how UMan, our mobile manipulation platform, obtains a kinematic model of an unknown object. The model then enables the robot to perform purposeful manipulation of that object. Our algorithm is extremely robust, and does not require prior knowledge of the object; it is insensitive to lighting, texture, color, specularities, and is computationally highly efficient."
Now, what's next on the researchers' agenda? They "will focus on teaching UMan to operate different types of machines, including doorknobs and light switches, and work on taking UMan’s manipulation skills into three dimensions. 'Once robots learn to combine movement, perception and the manipulation of objects, they will be able to perform meaningful work in environments that are unstructured and constantly changing,' says Katz. 'At that point, we will have robots that can explore new planets and clean houses in a flexible way.'"
Sources: University of Massachusetts Amherst news release, June 3, 2008; and various websites
You'll find related stories by following the links below.