Teaching RoboDog New Tricks

Back in 2008, it was easy to make fun of DARPA's BigDog quadruped robot that was learning to walk.  And last fall, the smaller robot LittleDog looked laughably clumsy trying to climb stairs and get over obstacles.

But over the past eight months at the Computational Learning and Motor Control Lab at USC, LittleDog has been developing new skills, and nobody's laughing now It's pretty darned adept at "safe, fast, autonomous locomotion over challenging terrain." Beware the RoboDog Rebellion.

- via gizmodo

From the Upcoming ueue, submitted by Marilyn Terrell.