Miss Georgia tripped in the final round of the 2015 Miss America Pageant. Jennifer Lawrence stumbled on her way to accept an Oscar. Even rock stars, world leaders, and presidential candidates have fallen in front of the crowd or completely off stage. And robots can too.
You may remember the disastrous (and hilarious!) videos from the 2015 DARPA Robotics Challenge of robots falling over themselves in a competition. As funny as that is, it presents a real world issue that engineers must overcome: how do we keep robots from falling over? And if that doesn’t work, how do we protect these multimillion dollar machines from damaging themselves on the way down?
Ph.D. graduate Sehoon Ha and Professor Karen Liu of Georgia Tech may have the answer. According to ScienceDaily, they have identified a way to teach robots how to fall with grace and without incurring serious damage. They developed a new algorithm that tells a robot how to react to a wide variety of falls—from a single step to recover from a gentle nudge, to a rolling motion that breaks a high-speed fall.
“A fall can potentially cause detrimental damage to the robot and enormous cost to repair,” says Ha, who graduated in summer 2015 and is now a postdoctoral associate at Disney Research Pittsburgh in Pennsylvania. “We believe robots can learn how to fall safely.”
Which is good news for the bystanders filming a falling robot that might get hit by, well, a falling robot! Thanks to the team’s research, robots can now minimize the damage or injury they might cause to themselves or others while falling by learning the best sequence of movements to slow their momentum. And we’ve got the video to prove it!
The research grew out of Liu’s study of cats and how their bodies modify when falling down. Liu knew from that work that one of the most important factors in a fall is the angle of the landing. She also knew that a well-designed robot has the “brain” to compute a softer landing, but doesn’t have the practical know-how to achieve that landing. Especially in the middle of a fall.
“[The robots] didn’t have the hardware to move quickly enough like a cat,” says Liu. “Our new planning algorithm takes into account the hardware constraints and the capabilities of the robot, and suggests a sequence of contacts so the robot gradually can slow itself down.”
So now robots can add cat-like reflexes to their arsenal of abilities that will cause them to rise up and take over humanity …but don’t tell the robots we said that!
Photo credit: IEEE Spectrum