Cornell robot is "conscious," adapts to injury

Oh great. As if the potent, robotic cocktail of self-replication, self-awareness, and wireless power weren't bad enough, along comes Cornell University with a robot capable of not only discovering its own nature (something we can't even do) but then adapts to overcome injury. This four-legged robot starts out knowing only what parts it has, not where they are or how to use them for locomotion. It applies a scientific method of theory and experimentation to develop computer models and ultimately, a set of commands to turn its motors for that first cautious step. Even when researchers remove part of the toddling robot's leg, the little guy still figures out a way to limp forward. Cornell scientists go so far as to say that the robot is "conscious," albeit on a primitive level since it thinks to itself, "what would happen if I do this?" Yeah, that's a stretch, but a step closer to our doom nevertheless. Although the robot used to demonstrate these cognitive gymnastics is quite simple, the algorithm could be used to build more complex robots for say, space exploration or defending itself while standing in line for a PS3. Well, at least The Robots don't currently have a place to coalesce under a common roof of intellectual ferment... oh God, no! Still, any robot which drinks puddle water can't be too bright, eh?

[Rohit Kapur @ Nov 17th 2006 10:09AM]

I, for one, welcome the idiots who'll make an overlord joke on this post.

[Dom @ Nov 17th 2006 10:50AM]

But surely you're one of th... wait. Ahhhhh, the ironing.

[Todd @ Nov 17th 2006 10:24AM]

With the post, the doomsday clock just clicked forward a couple of seconds.

[shirizaki @ Nov 17th 2006 10:27AM]

Yeah, but do they have weak points I cna attack for massive damage? THE FUTURE DEPENDS ON IT!!!!!!

[Phree @ Nov 17th 2006 10:42AM]

.....and so it begins.

[R @ Nov 17th 2006 10:44AM]

After a case of beer, then it will go back to being UNconscious.

[Christian Martin @ Nov 17th 2006 10:46AM]

Looks like a Protoss contraption to me.

[Alex C @ Nov 17th 2006 10:50AM]

I wrote a small online article explaining in layman's terms the principles of this method of "consciousness". It really is the way to go. To summarise, you supply the machine with a variety of tools and an objective (much like a real life organism). By trying out its "tools" in different ways it discovers the most effective way to attain its objective. In the same way that a baby learns to walk.

Once you add the ability for the machines to communicate, they can optimise their methods and increase the speed of learning because they can all share the same knowledge base. I'm sure you'll be hearing a lot mroe from these guys.

[Matt B @ Nov 17th 2006 11:21AM]

What happens when I cut it straight down the middle with my axe? Will it become 2 beasts?

[Hardcore @ Nov 17th 2006 11:22AM]

We're boned.

[Whatsizface @ Nov 17th 2006 11:23AM]

I for one welcome these new robotic toddling overlords.

Wait! He's already here and sleeping in the crib in the next room!

[boynamedsue @ Nov 17th 2006 11:36AM]

I for one curse Rohit Kapur for beating me to that joke.

pussies bow to _______ overlords.

[mespinosam @ Nov 17th 2006 11:51AM]

Y'all calm down before applying turing test on this bot ... but in some way its only the beginning of something ...

[zenprime @ Nov 17th 2006 12:04PM]

I'm sure one day Engadget will be heralded for its efforts in tracking and publicizing the various breakthroughs that will lead to the birth of our Future Robotic Overlords. Asskissers.

[eddie @ Nov 17th 2006 12:13PM]

Please robot, I am the only one in this house, I don't have a family.

[Greg S @ Nov 17th 2006 12:44PM]

Skynet became self-aware at 2:14am August 7, 1997.

I guess it'll be about 10 years late.

[Kevin @ Nov 17th 2006 1:58PM]

Digging the Starcraft joke....

Does anyone else remember that??

[crazylink @ Nov 17th 2006 3:00PM]

"Ahhh! Woooh! What's happening? Who am I? Why am I here? What's my purpose in life? What do I mean by who am I? Okay okay, calm down calm down get a grip now. Ooh, this is an interesting sensation. What is it? Its a sort of tingling in my... well I suppose I better start finding names for things. Lets call it a... tail! Yeah! Tail! And hey, what's this roaring sound, whooshing past what I'm suddenly gonna call my head? Wind! Is that a good name? It'll do. Yeah, this is really exciting. I'm dizzy with anticipation! Or is it the wind? There's an awful lot of that now isn't it? And what's this thing coming toward me very fast? So big and flat and round, it needs a big wide sounding name like 'Ow', 'Ownge', 'Round', 'Ground'! That's it! Ground! Ha! I wonder if it'll be friends with me? Hello Ground!"

[alex @ Nov 17th 2006 3:22PM]

It's not drinking, it's waiting for you to step in the puddle then.... ZAP! One down, 5.999999 billion to go!

[Thomas Ricker @ Nov 17th 2006 4:12PM]

crazylink,

Thanks, that comment made the perfect end to my week.

Thomas

[Crazylink @ Nov 17th 2006 7:29PM]

I'm glad I could help.

[S.D. @ Nov 17th 2006 5:48PM]

Wow, reminds me of a Stargate Replicator!

[Mr. Picklesworth @ Nov 17th 2006 7:53PM]

It's cool, but it is not conscious.
It looks, but does it /see/?

Still, this is an incredible accomplishment in robotics and AI. Just don't act it's a living thing, hm?
I'd rather machines remain completely disposable. I like sending them to near-death situations, and I don't want them winning my heart and causing me to risk my own life out of pity. (There's probably a dumb movie which I could say I'm referring to here!)
What would have happened if those Mars rovers were sentient? They would have taken over their carriers, turned around and crashed into NASA in no time!

Er, anyway... I am really looking forward to seeing what can be done with this technology. It's coming along perfectly, and I really appreciate that the first robot to be like this is not built to look like a human. So far, they clearly do not do the above, and hopefully they never will.

Hehe, nice post, Crazylink.

[Othello @ Nov 17th 2006 11:46PM]

Hey, for once I beat you guys to an article :o

[Systema @ Nov 17th 2006 11:52PM]

Picklesworth, if it looks and can interpret what it is looking at, then it sees.
Humans should start realizing that they're nothing special.

Signed,
An Angry Martian.

[Aaron @ Nov 18th 2006 7:08AM]

Wow, this is exciting. I can't wait to see where this goes.

Although I don't think it's particularly new. I remember reading a few years ago about a humanoid robot that they built with eyes and ears and upper body movement (I don't think it could walk) and then they taught the damn thing to juggle, by actually teaching it. The idea was to make it learn and in the end have the overall "intelligence" of a two year old child.

Still though, very interesting. The first step in making "living" machines.

[jessew @ Nov 22nd 2006 12:02AM]

Really Mr. Picklesworth, do you think a(dare I say any?) government body views a single person as any thing other than a dispensable object? The trend of the time is a resounding 'no'

If a man didn't procreate in the normal sense but invented the most compelling AI you have ever
seen, would it be his child? no? what if it took him 16 years to make? Hard to say about the future...

[T-1000 @ Feb 1st 2007 1:54PM]

I think this is cool I want one of these things!!!!!