Scientists have created a robot that has taught itself how to accurately hit a target with a bow and arrow, and mankind is one step closer to total domination. Okay, maybe that’s a bit of a stretch, but why do we have to teach them so much? You know that’s going to come back and bite us in the butt, right?
The robot, named iCub, was designed by researchers at the Italian Institute of Technology. Oblivious to political correctness, they dressed the little humanoid in a Native American headdress and thanks to a complicated computer algorithm, the robot learned from its missed shots until it hit the bull’s-eye.
Researchers say that choosing the bow and arrow has nothing to do with arming little humanoids to take over the planet, but was picked purely for its inherent and obvious reward, and simultaneous marriage of motor control with image processing. Why do I have a hard time believing that? Too many Terminator movies, I guess.
The first iteration of iCub hit the bull’s-eye, standing three and a half meters from the target, in eight attempts. While not exactly Olympic archery material, it is pretty impressive. iCub uses a a learning algorithm called ARCHER, or Augmented Reward Chained Regression, which implements a camera to process the bull’s-eye image, and his previously fluffed attempts, to figure out the perfect angle, force and trajectory to make the winning shot.
Earlier this year, the researchers at the Italian Technology Institute also taught a Barrett WAM 7 robotic arm to flip pancakes all by itself. While it took 50 attempts to get it right, they obviously have tweaked the model a bit with iCub.
iCub will present his archery-mastering skills at the Humanoids 2010 conference this December in Tennessee, along with a passenger carrying a biped, musical conducting robots, a Mini-Humanoid Pianist and a robot that can play table tennis.
Wait, there’s an entire conference dedicated to Humanoids, and it’s in December? For some reason I can’t help but think of December, 2012. You know, when the world is supposed to end. Could this Humanoid conference be the catalyst? No, of course not. I’m kidding. Still, one can’t help but wonder where it’s all going to stop. First a bow and arrow, then a shotgun, maybe a flamethrower? Research presses on, though, and continues to test the limits of artificial intelligence. Mark my words, good reader. It won’t be long before you hear a robot mumble, “I’ll be back,” and boy will it ever.