Adam Coates of Stanford presented at the IEEE RAS meeting. He has his colleagues, including Pieter Abbeel and Andrew Y. Ng, used machine learning to enable a 2-processor computer to quickly learn complicated flight patterns of large model helicopters controlled by a skilled pilot. The model helicopter pilot took 20 years of learning to perfect some of the complicated and aggressive flights and the computer was able to learn it in a few hours. Thereafter, the computer can repeatedly control the model helicopters thru the same flight pattern. Adam indicated that they looking into applying these techniques for other types of robot learning tasks.
Copyright (c) 2008 by Waiming Mok