Dealing with obstacles is an inevitable part of life, and it looks like robots may be surprisingly adept at applying creativity to the challenge.
Researchers from Carnegie Mellon University developed software that not only helped a robot deal efficiently with clutter but also revealed considerable creativity in solving problems. Their new study is due to be presented Thursday at the IEEE International Conference on Robotics and Automation in Sweden.
A research team led by Siddhartha Srinivasa, CMU associate professor of robotics, challenged HERB -- his lab's two-armed mobile robot -- with a pile of clutter.
What they saw from the robot surprised them. In one case, the robot, officially named the Home Exploring Robot Butler, used the crook of its arm to cradle an object to be moved.
“We never taught it that,” Srinivasa said.
“It was exploiting sort of superhuman capabilities,” Srinivasa added. “The robot’s wrist has a 270-degree range, which led to behaviors we didn’t expect. Sometimes, we’re blinded by our own anthropomorphism.”
The software was also tested on NASA’s K-REX robot, which is being designed to traverse the lunar surface. While HERB focused on clutter typical of a home, KRex used the software to find traversable paths across an obstacle-filled landscape while pushing an object.
It's long been known that robots are adept at “pick-and-place” processes, whereby they pick up an object in a specified place and put it down in another. That's great in relatively clutter-free locations like factory production lines, but not so much in homes -- or on distant planets.
After all, when you reach for a milk carton in a refrigerator, you don’t necessarily pick up and move every other item out of the way; rather, you might move an item or two but then shove others aside.
The new software automatically finds a balance between those two strategies based on the robot’s progress on its task. The robot is programmed to understand the basic physics of its world, so it has some idea of what can be pushed, lifted or stepped on. It can also be taught to pay attention to items that might be valuable or delicate.
Currently, CMU's algorithm forges ahead full-bore with any plan it makes, but work is under way to help it make corrections along the way.
Sign up for CIO Asia eNewsletters.