Robots do a lot for us these days, so it’s no surprise to learn that even they feel run down, meaning sometimes they break a leg or lose an arm. According to new research at Pierre and Marie Curie University in Paris, robots that are broken in motion can now quickly heal themselves by tapping into experiences from simulated lives.
Though it sounds like science fiction, it’s anything but. In experiments, a six-legged robot adjusted itself in little more than a minute to keep walking, even if two of its legs were damaged, broken, or missing. A robotic arm learned to place an object in the correct place, even with several broken motors or joints.
“One thing we were surprised by was the extent of damage to which the robots could quickly adapt to,” roboticist Jean-Baptiste Mouret told Live Science. “We subjected these robots to all sorts of abuse, and they always found a way to keep working.”
A damaged robot. Image via Live Science.
The idea behind these experiments comes from nature. Animals are often able to adapt rapidly from injuries. For example, three-legged dogs can still run to catch a Frisbee, and humans can figure out how to walk despite something such as a sprained ankle.
“If we send in robots to find survivors after an earthquake, or to put our forest fires, or to shut down a nuclear plant in crisis like Fukushima, we need them to be able to keep working if they become damaged,” Mouret said. “In such situations, every second counts, and robots are likely to become damaged because these environments are very unpredictable and hostile. Even in less extreme cases, such as in-home robot assistants that help the elderly or sick, we want robots to keep performing their important tasks even if some of their parts break.”
Scientists have now developed a trial-and-error program that enables robots to adapt to damage in less than two minutes, without a suite of sensors to diagnose itself. They reasoned that animals don’t learn how to recover injuries from the start, but have intuitions about different ways to behave, allowing them to intelligently select different behaviors to try out, and after, they choose one that works in spite of the injury. The self-healing robots were made to do the same.
Before a robot is deployed in this new strategy, the scientists develop a computer simulation to map out thousands of different motions it can take, and predict which patterns of actions are likely to work despite damage.
“We do not pre-compute anything like 'find a gait that works if a leg is missing,'” Mouret said. “What we do with the simulator is simply to say 'find as many different ways to walk as you can.'”
Researchers suggest this strategy could help robots adapt to unforeseen circumstances and new environments. Some potential applications include robots that can help rescue workers, and robots that can serve as personal assistants.
Story via Live Science.
Advertisement
Learn more about Electronic Products Magazine
