Truemors
is reporting from
Member
NP Rank:
NP Rank:
Stanford's 'autonomous' helicopters teach themselves to fly
Share: 
by Erik Larson | September 3, 2008 at 03:06 pm
157 views | 5 Recommendations | 5 comments
Scratch another one off the list of "things only humans can do". This is not even a full artificial "intelligence"; it just learns how to compute data and algorithyms to fly a remote control helicopter, beaming instructions by radio signal 20 times a second. However, that doesn't mean it's any less fun, or that there is any less need for people to buy remote control helicopters and fly them for fun! It does mean these things will save lives fighting fires and searching for land mines- in addition to whatever other military and crowd control scenarios are dreamed up that aren't "fit to print".
Computer Science Professor Andrew Ng (center) and his graduate students Pieter Abbeel (left) and Adam Coates have developed an artificial intelligence system that enables these helicopters to perform difficult aerobatic stunts on the their own. The “autonomous” helicopters teach themselves to fly by watching the maneuvers of a radio control helicopter flown by a human pilot.
------
There is interest in using autonomous helicopters to search for land mines in war-torn areas or to map out the hot spots of California wildfires in real time, allowing firefighters to quickly move toward or away from them. Firefighters now must often act on information that is several hours old, Abbeel said.
What is NowPublic?
NowPublic lets people work together to cover news events around the world.
Most RecentMost Recommended Comments (5)
at 15:40 on September 3rd, 2008
I swear robots are going to take us out and take over the world one day.
at 22:30 on September 3rd, 2008
What is friendly AI?
Revenge of the Nerds, huh? Given that trends in solar technology mean there will be no need for resource conflicts, and that an intelligence not limited by evolution or biology will be much smarter than human, i'm optimistic they'll be more interested in helping us solve our stupid, simple problems, just as they are now.
The day a robot can grab its crotch and offend more people with a rendition of the national anthem than Rosanne Barr, we might be in trouble.
And could an AI come up with a youtube classic like this, and get over 8 million hits in less than a year? Britney Fan Crying (Chris Crocker)
at 22:35 on September 3rd, 2008
I think we are already in trouble, the first Cyborg have been build with Rats brains. The First Computers able to learn and evolve on their own are already in existence. Soon the First Human with Cyborg implant and or Micro chip implant will be officially known to the public and the Nano Robots will perform micro surgery in ways we could not even imagine only two decades ago. The first Androids will be in operation with in the next decade and we may have created our successors on Earth.
at 22:20 on September 4th, 2008
thanks, Paschen, i agree this is a significant story, and part of it is how news like this is increasingly common; cyborg rats as you say- and with human/computer interface coming. I think of it as humanity transcending biological limits. It's a logical next step and one that mirrors the same trend towards increasing information, organization and complexity that can be seen in the transformation of this Universe from the Big Bang to the present Network Age. We'll make great pets. Actually, i do think the trend of computers being our friends will continue. Information technology has contributed hundreds of billions to economic growth and is enriching nations around the world, with knowledge, wealth and opportunities. In this decade, computers will translate in real time cheap so we can talk to anyone in the world.
at 22:36 on September 3rd, 2008
Erik Larson, I like this story. It's good stuff.