We tend to use what Goldberg calls “robust grips.” These are secure grips that work even if we run into problems with perception, control or physics. Millions of years of evolution provided brains and bodies with ways to adapt to all three challenges. “My dog Rosie is pretty good at grasping anything in our house,” says Goldberg. Goldbergĭespite these challenges, people and other animals grasp things all the time - with hands, tentacles, tails and mouths. Ken Goldberg’s dog, Rosie, is great at grabbing onto toys with her mouth. People and animals have evolved effective ways to grasp all kinds of things. Goldberg says, “If you push it the same way three times, the pencil usually ends up in a different place.” Very tiny bumps on the floor or the pencil may change the motion. Put it back in its starting place and try again. To see why, put a pencil on the floor, then give it a big push. But on small scales, this can be unpredictable. To grasp something, you must understand how that object could shift when you touch it. Likewise, a robot’s cameras and sensors won’t always be in perfect sync with its moving “hand.” If the robot can’t tell exactly where its hand is, it could miss something or drop it. You likely won’t be able to touch the exact same spot on your nose every single time. Try to do it fast!” Then try a few more times. To test yourself, Goldberg says, “Reach out, then touch your nose. People are good at this, but not perfect. This is your ability to move your hand accurately. But robots still get confused by anything “shiny or transparent,” he notes. Cameras and sensors that measure distance have gotten much better at this in recent years. That’s the ability to see an object and figure out where it is in space.
He’s discovered that robots (and clumsy humans) face three challenges in grabbing an object.
Maybe he’d figure out the secret to holding onto things. Perhaps, he ponders, that’s why he wound up studying robotic grasping. If you threw me a ball, I was sure to drop it,” he says. Goldberg has something in common with robots. We’ll also learn about the brilliant work engineers and computer scientists are doing to design new AI that should help robots up their game. For each one, we’ll find out why the task is actually so hard. Let’s take a look at several tasks that are easy for kids but not for robots. You just don’t notice it because you do it without thinking. As you walk around your house or pick up and move a chess piece, your brain is performing incredible feats of calculation and coordination. It turns out that the tasks we find easy aren’t really “easy” at all. Though computers have advanced by leaps and bounds since then, babies and kids still beat machines at these types of tasks. It’s “difficult or impossible” for a computer to match a one-year-old baby’s skills in these areas, he wrote. Meanwhile, many tasks that come easily to us - like moving around, seeing or grasping things - are quite hard to program. Back in 1988, he wrote a book that noted how reasoning tasks that seem hard to people are fairly easy to program into computers. Hans Moravec is a roboticist at Carnegie Mellon University in Pittsburgh, Penn., who also writes about AI and the future. Let’s learn about artificial intelligence Yet a robot has trouble picking up an actual chess piece. A computer can easily defeat a human grandmaster at the game of chess by coming up with better moves. He’s an engineer and artificial intelligence (AI) expert at the University of California, Berkeley. But when it tries to grab a knight, it knocks down a row of pawns. Its computer brain calculates a winning move in a fraction of a second. You reach out and push your queen forward. Finally, you see a move that looks pretty good. You’re sitting across from a robot, staring at a chess board.
0 kommentar(er)
