Miss Georgia tripped in the final round of the 2015 Miss America Pageant. Jennifer Lawrence stumbled on the way to accepting her Oscar. Rock stars, world leaders, and presidential candidates have all fallen in front of crowds.
And robots can too.
Now, researchers have identified a way to help robots fall with grace — and without serious damage. This is important now that costly robots have become more common in manufacturing and are sought for health care or domestic tasks.
Georgia Tech Ph.D. graduate Sehoon Ha and Professor Karen Liu have developed a new algorithm that tells a robot how to react to a wide variety of falls — from taking a single step to recover from a gentle nudge, to going into a rolling motion to break a high-speed fall. By learning the best sequence of movements to slow their momentum, robots can minimize the damage or injury they might cause to themselves or others while falling. The algorithm has been validated in physics simulation and experimentally tested on a BioloidGP humanoid robot.
“A fall can potentially cause damage to the robot and enormous cost to repair,” said Ha, now a postdoctoral associate at Disney Research Pittsburgh. “We believe robots can learn how to fall safely. Our work unified existing research about how to teach robots to fall by giving them a tool to automatically determine the total number of contacts (how many hands shoved it, for example), the order of contacts, and the position and timing of those contacts. All of that impacts the potential of a fall and changes the robot’s response.”
The latest finding builds upon Liu’s previous research that studied how cats modify their bodies in the midst of a fall. Liu knew from that work that one of the most important factors in a fall is the angle of the landing. Armed with that information, the two researchers used the robots’ silicon brains to optimize the sequence of motions that take place during a fall.
“From previous work, we knew a robot had the computational know-how to achieve a softer landing, but it didn’t have the hardware to move quickly enough like a cat,” said Liu, an associate professor in Georgia Tech’s School of Interactive Computing. “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.”
The research was presented at the IEEE/RSJ International Conference on Intelligent Robots and Systems in Hamburg, Germany.
— Jason Maderer