Bow to our new Robot Overlords!

ThatGuyFromThatPlace · Post by **ThatGuyFromThatPlace** » 2006-11-17 12:06am

WASHINGTON (Reuters) - It's an achievement that inspires notions of robots with consciousness and independent minds.

Scientists said on Thursday they created a brainy, four-legged robot resembling a starfish that can sense damage to its body and, on its own, think up a way to recover.

Researchers Hod Lipson and Victor Zykov of Cornell University and Josh Bongard of the University of Vermont made a robot that observed its own motion using built-in sensors in its joints and then generated its own concept of itself, or at least its physical structure, in its internal computer.
al model of itself to figure out how to walk on its four legs and eight motorized joints.

"In the beginning, the robot starts off and does not know what it looks like. You look at it, and you see that it's a four-legged machine. But the robot itself doesn't know that. All it knows is that it could be a snake, it could be a tree, it could have six legs," Lipson said in an interview.

Lipson said the robot used various movements of its joints, first to generate hypotheses and then to formulate an accurate conception of itself.

The researchers then tested the robot's ability to adapt to new situations -- in this case injury -- by shortening one of its legs. "The robot knows something's wrong," Lipson said.

Darth Servo · Post by **Darth Servo** » 2006-11-17 12:14am

And the time until the Trekkies start claiming Star Trek is responsible for this in 10...9...8...7...

Lord of the Abyss · Post by **Lord of the Abyss** » 2006-11-17 01:34am

Darth Servo wrote:And the time until the Trekkies start claiming Star Trek is responsible for this in 10...9...8...7...

And the time until Warsies complain about Trekkies 10...whoops, already happened !

Zor · Post by **Zor** » 2006-11-17 02:24am

Cool, i think the military would be interested in this.

Zor

JointStrikeFighter · Post by **JointStrikeFighter** » 2006-11-17 04:34am

I for one welcome our new mechanical overlords.

ThatGuyFromThatPlace · Post by **ThatGuyFromThatPlace** » 2006-11-17 11:01am

I was just thinking that should have been the title instead of " Bow to our new Robot Overlords!"

Darth Servo · Post by **Darth Servo** » 2006-11-17 01:27pm

Lord of the Abyss wrote:
Darth Servo wrote:And the time until the Trekkies start claiming Star Trek is responsible for this in 10...9...8...7...
And the time until Warsies complain about Trekkies 10...whoops, already happened !

Whats wrong with complaining about stupidity and lies?

OmegaGuy · Post by **OmegaGuy** » 2006-11-17 01:53pm

Darth Servo wrote:
Lord of the Abyss wrote:
Darth Servo wrote:And the time until the Trekkies start claiming Star Trek is responsible for this in 10...9...8...7...
And the time until Warsies complain about Trekkies 10...whoops, already happened !
Whats wrong with complaining about stupidity and lies?

Well for one thing, they haven't happened yet.

Magus · Post by **Magus** » 2006-11-17 02:08pm

Hmm...I wonder what the non-military applications of this are? Self-repairing, self-aware robots...

Tasoth · Post by **Tasoth** » 2006-11-17 03:22pm

Magus wrote:Hmm...I wonder what the non-military applications of this are? Self-repairing, self-aware robots...

Friends? Yeah, I'm a romantic. I wonder what would happen if you plugged this into one of those neuronets that are freakishly creative.

SirNitram · Post by **SirNitram** » 2006-11-17 05:21pm

OmegaGuy wrote:
Darth Servo wrote:
Lord of the Abyss wrote: And the time until Warsies complain about Trekkies 10...whoops, already happened !
Whats wrong with complaining about stupidity and lies?
Well for one thing, they haven't happened yet.

I see your reading comprehension isn't as high as it could be. Servo is merely predicting they will, and in short order. Given Trekkies have already tried to claim Trek inspired all robotics and AI development(Hilarious with all the Asimov references), that's not inaccurate.

So shut the fuck up.

On the OP: Neat stuff.

Darth Servo · Post by **Darth Servo** » 2006-11-18 01:40am

OmegaGuy wrote:
Darth Servo wrote:
Lord of the Abyss wrote: And the time until Warsies complain about Trekkies 10...whoops, already happened !
Whats wrong with complaining about stupidity and lies?
Well for one thing, they haven't happened yet.

I guess you've never seen the crap they spew about cell phones, CDs and desktop computers?

DrMckay · Post by **DrMckay** » 2006-11-18 02:14am

how's about making a kill-switch in each self-aware robot, like an ounce or two of C-4. Just in case...

Stark · Post by **Stark** » 2006-11-18 02:17am

In case real engineers are as stupid as Hollywood-movie engineers?

Then again, the US can't design an electronic voting machine. Imagine Diebold humaniform robots...

Lord Poe · Post by **Lord Poe** » 2006-11-18 05:22am

Er... are we sure the robot "thinks" and "decides" how to best move itself? Why wouldn't it just access the possible locomotive choices it must have in its brain, until it come across one that works?

Also, someone must "tell" or motivate it to move. Why would it move on its own if it didn't have a live or preprogrammed reason to do so?

Obviously, if there's a defect in its movement template, it would be "aware" of this problem. My car is "aware" of it being low on fuel.

Magus · Post by **Magus** » 2006-11-18 09:54am

Lord Poe wrote:Obviously, if there's a defect in its movement template, it would be "aware" of this problem. My car is "aware" of it being low on fuel.

Not quite...your car simply has a gauge that tells you when it's low on gas. From what I understand, what separates this robot from others is that, where others will attempt to complete their movement template despite mechanical error (probably falling over in the process), this one will adjust its movement to compensate.

In other words, I think it can modify its own programming to better achieve a programmed criteria of objectives - moving and remaining upright, for example.

Lord Poe · Post by **Lord Poe** » 2006-11-18 12:00pm

Magus wrote:Not quite...your car simply has a gauge that tells you when it's low on gas. From what I understand, what separates this robot from others is that, where others will attempt to complete their movement template despite mechanical error (probably falling over in the process), this one will adjust its movement to compensate.

In other words, I think it can modify its own programming to better achieve a programmed criteria of objectives - moving and remaining upright, for example.

Modify its own programming? I'm not sure about that. Again, wouldn't it just shuffle through its possible movement templates until it found one that allowed it to move? Is there something in its brain saying not to use the default template if the locomotion protocols tell it the movement would be abnornal?

I just don't see this as any kind of "breakthru" in AI.

ThatGuyFromThatPlace · Post by **ThatGuyFromThatPlace** » 2006-11-18 01:43pm

it doesn't have a 'default' move template. from what I understand what it has is a series of commands for controlling the various servos that make up its means of locomotion and it triggeres them in some fashion until it finds the sequence that lets it move most efficiently. the, if something changes, it alters the movement pattern to compensate.
And it is pretty revolutionary.

Magus · Post by **Magus** » 2006-11-18 03:04pm

ThatGuyFromThatPlace wrote:it doesn't have a 'default' move template. from what I understand what it has is a series of commands for controlling the various servos that make up its means of locomotion and it triggeres them in some fashion until it finds the sequence that lets it move most efficiently. the, if something changes, it alters the movement pattern to compensate.
And it is pretty revolutionary.

Exactly - achieving complexity by programming simplistically

Admiral Valdemar · Post by **Admiral Valdemar** » 2006-11-20 11:32am

In other words, it's bottom-up programming of an AI, rather than top-down which means coding a response to every possible situation i.e. nigh impossible. The use of heuristic programming for AIs and neural nets means robots that are able to evolve to adapt to new situations, much like the Seven Dwarves.