September 1st, 2008 by ketyung
Some computer experts at Stanford University have designed an intelligent system and install them on a troupe of autonomous helicopters, which let them fly better than the RC pilot who trained them.
These robotic helicopters are equipped with intelligent self-learning system which allows them to quickly analyze data feed to them from the ground and from other copters. They send about 20 instructions in every second to the units in flight.
The intelligent computer system allows the helicopters to analyze the real reaction of a human RC pilot, such as on how they respond when going against the wind. This makes the Stanford’s copters able to watch, learn and adjust itself to changes in the air more fluidly. The helicopters used in their experiment are simply those ordinary ones which they bought it from stores that sell RC gadgets. With their computer systems that managed to make a big difference from previously produced autonomous copters which had high tendency to get a crash.
Those folks in the team are Professor Andrew Ng, graduate students Pieter Abbeel, Adam Coates, Timothy Hunter, Morgan Quigley, and expert remote controller Garett Oku.
via [dvice]
robotic helicopters,RC helicopters,self-learning helicopters
Technorati Tags: robotic helicopters, RC helicopters, self-learning helicopters
Stumble it!
September 2nd, 2008 at 6:30 am
[...] TechChee If you enjoyed this post, make sure you subscribe to our RSS [...]