Robots Master Motor Skills by Trial and Error

#Robots

Fri, May 22nd, 2015 11:00 by capnasty NEWS

By mimicking the process in which humans learn, researchers at UC Berkeley have developed algorithms that allow robots to master their motor skills by trial and error. The idea is that when the robot is faced with a new task, it won't need to be programmed, but will instead figure it out on its own.

“What we’re reporting on here is a new approach to empowering a robot to learn,” said Professor Pieter Abbeel of UC Berkeley’s Department of Electrical Engineering and Computer Sciences. “The key is that when a robot is faced with something new, we won’t have to reprogram it. The exact same software, which encodes how the robot can learn, was used to allow the robot to learn all the different tasks we gave it.”

  1769

 

You may also be interested in:

“It will be very disruptive and very quick.”
"The [robot] could keep children [...] occupied for 'a couple of hours' without adult supervision."
Robotic Guide Dog Uses Microsoft Kinect Controller to Navigate
“Robots of reality are starting to get to the point where they have similar capabilities to the robots of our imagination.”
"Gathering human perspective on moral decision made by machine intelligence."