Quantcast

New Algorithms Give Household Robots A Helping Hand

February 25, 2013
Image Credit: Willow Garage

Lee Rannals for redOrbit.com — Your Universe Online

With a few adjustments through algorithms, MIT researchers found a way to give household robots an extra hand.

Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory will be presenting two papers at this year’s IEEE International Conference on Robotics and Automation, showing robots could use algorithms to compensate for physical shortcomings.

One paper describes an algorithm that enables a robot to push an object across a table so it hangs off the edge, making it easier to grab with its hand. The other paper shows how a two-armed robot can use one of its hands to steady an object set in place by the other.

For the first paper, Jennifer Barry, a PhD student in the group, had to find a way to represent the physical properties of the object that needed to be pushed and show how it would respond to different forces once pushed in a particular direction.

“This allows us to focus the search on interesting parts of the space rather than simply flailing around in 16 dimensions,” she said in a statement. Finally, after her modification of the motion-planning algorithm, she had to “make sure that the theoretical guarantees of the planner still hold,” she says.

With the other algorithm, Annie Holladay, an MIT senior majoring in electrical engineering and computer science, had to identify the paths that introduce collisions and seals them off.

“I might look for a place for the other hand that will block bad paths and kind of funnel the object into the path that I want,” Holladay explained.

Annie had to find a simple method of representing the physical properties of the object, and also create an algorithm that can handle cases in which the robot is setting an object on a table, but it sticks to the robot’s grippers before hand. With this algorithm, the robot is able to use its free gripper to prevent the object from sliding as it drops from the other gripper.

“You see a lot of demos where a robot might do something like slide plates, but it´s usually hard-coded for the demo: The robot knows that at this point, it needs to do this action for this particular thing,” says Kaijen Hsiao, a research scientist and manager at Willow Garage, the company that manufactures the PR2, the robot used for the study.

Hsiao said the team’s research is the framework for incorporating behaviors that relate to a general motion-planning problem. She said this is a very difficult thing to do, because it is very high-dimensional.

“I think it´s really important research, and it´s very novel,” Hsiao added.

There is constant research going on in the field of robotics, and scientists look to everything for inspiration to create the next-generation butlers and even cockroaches. Recently, University of Michigan researchers wrote in the journal Biological Cybernetics about how they found that running cockroaches have inspired robotics.

When running, cockroaches can recover from being pushed sideways, even before their nervous system kicks in to let their legs know what needs to happen. Through kinematic phase analysis, the scientists were able to get a better picture on how cockroaches are able to do this. Shai Revzen, an assistant professor of electrical engineering and computer science, said the team’s research could be valuable in the biomedical community for human gait analysis.


Source: Lee Rannals for redOrbit.com – Your Universe Online



comments powered by Disqus