A Thousand Kilobots Self-Assemble Into Complex Shapes 56
An anonymous reader writes "Researchers at Harvard's Self-Organizing Systems Research Group—describe their thousand-robot swarm in a paper published today in Science (they actually built 1024 robots). In the past, researchers have only been able to program at most a couple hundred robots to work together. Now, these researchers have programmed the biggest robot swarm yet. Alone, the simple little robot can't do much, but working with 1,000 or more like-minded fellow bots, it becomes part of a swarm that can self-assemble into any two-dimensional shape. These are some of the first steps toward creating huge herds of tiny robots that form larger structures—including bigger robots."
Re: (Score:2)
Well, for starter, they typo'd "kilbots"
Re: (Score:2)
Well, for starter, they typo'd "kilbots"
Working our way up to Gray Goo armageddon.
They couldn't name them killbots, because (1) Killbots are illegal due to the anti-killbot act, and (2) The name Killbot (TM) has already been trademarked by a large overseas manufacturer who so far declindes to comment on questions about what the upcoming product will be.
Re: (Score:2)
Is it a thousand killbots with a typo, or one megabots?
Re: (Score:2)
Neither. It's a kilo-exaggeration (what happens when you exaggerate by a factor of 1000). Also, the individual bots are called Kilobots, because the creators like confusing names.
Re: (Score:2)
I also misread at first and needed a double take. If it was 1024 killbots I'd be rather worried.
If they're Futurama killbots we can just throw wave after wave of soliders and police into them until they exceed their kill limit safeties.
If they were more like Terminator killbots, the world would be screwed.
But since they're kilobots rather than killbots, having a kilo of kilobots sounds like fun.
Re: (Score:2)
If it takes 1000 of these little guys to make a composite bot that is actually interesting, then they are in fact millibots.
Which means they're only off by six orders of magnitude.
Re: (Score:2)
Killbots (Score:3)
I mis-read and thought this said "A Thousand Killbots Self-Assemble Into Complex Shapes..."
Now THAT would be interesting!
Re: (Score:2)
You aren't alone. It reads like the beginning of a sci-fi novel.
Re: (Score:2)
Or it would be the world's most boring version of Voltron.
"I'll assemble a third of the lower left third of the pinky!"
"I'll assemble the upper tip of the ear!"
"I'll assemble the middle of the right elbow!"
plot from Crichton novel (Score:2)
Re: (Score:2)
Re: (Score:2)
lol, I think you'll still be ok for bit. It took them 11 hours to make the letter K. It would take them days just to cuss at you. To make something threatening would take... way longer than anyone is going hang around and keep them charged.
On the other hand...I'll take a set as cat toys. See how long they can hold the shape of a mouse :)
6-12 hours per formation....there is probably a good reason everyone uses virtual ones.
640k? (Score:5, Funny)
Re: (Score:2)
Imagine a Beowulf cluster...
Oh. Never mind. :-P
On a more serious note, this looks like the beginning of multicellular robotic life. Whee! How long until the grey goo?
Cheers,
Re: (Score:3)
My dear, we are the grey goo.
That's-a-lotta-bots-a (Score:2)
A Thousand Kilobots
So that's like, what, 1024000 bots?
Re: (Score:2)
Re: (Score:2)
More accurate would be to say that each bot is 1 millishape, and a kilobot is one shape.
A millibot would be a thousandth of a bot, not of a thousandth of a shape.
For example, if you join a football team then you are 1/nth of a football team, but we don't say generally say that makes you 1 nth of a human... (some might of course, but usually they have a pretty low opinion of football players and jocks in general)
Re: (Score:2)
Reminded me of Reedman... (Score:2)
What's the additional challenge here? (Score:2)
Other than the fact that fabricating that many 'bots is painful and expensive, what makes this different from The Game of Life (albeit with an algorithm that takes more than a couple of lines.) I just don't see how this is any different from running a simulation of robots forming "any 2-D shape"... what was learned by actually building them?
Re:What's the additional challenge here? (Score:5, Funny)
what was learned by actually building them?
How to successfully apply for grant money.
Re: (Score:3)
I think they're building these robots to solve the problem of how to make these robots. A pixel in a game of Life is easy to maintain -- it has an x,y coordinate and immediately knows all its neighbor's positions. A robot has to identify all its neighbors and then localize itself using infrared and communication time lags. That's a challenge. The only way to meet that challenge is to build the robots and figure out how to make them work.
Sweet (Score:2)
"Can't do much" (Score:2)
Alone, the simple little robot can't do much, but working with 1,000 or more like-minded fellow bots, it becomes part of a swarm that can self-assemble into any two-dimensional shape.
I'm pretty sure this means that working with 1,000 or more like-minded fellow bots, it still can't do much.
Expensive little suckers (Score:4, Informative)
Why is this better than simulation? (Score:3)
It's sort of cool, I guess, but I don't see the benefit of actually building physical robots rather than running a simulation. What has been achieved in the real world doesn't seem to have any practical application, even as an advertising gimmick or a work of sculpture.
I can't imagine sending out 100,000 of these gadget to do the half-time show at a football game, for example.
I didn't sense that this was just the beginning and that the same devices that self-assemble predetermined shapes could, with more advance software, harvest wheat or perform laser surgery.
When they reach the point where the simulated behavior actually has some real-world utility, THEN it makes sense to build them.
Why is this better than simulation? (Score:4, Informative)
The difference between theory and practice is
- in theory, there is no difference
- in practice, there is.
A simulation of self-assembling robots is theory.
An actual pile of 1,024 self-assembling robots is practice.
Less tritely, you have zero information about flaws in your simulation until you try to apply it to/in the real world. Your simulation is excellent at helping you identify logical flaws in your design. But if you fail to account for something (crosswinds [wikipedia.org], say), then your simulation simply won't help you find it.
It's that whole "unknown unknowns" thing, man.
Re: (Score:2)
Superbowl (Score:2)
Pretty sure if you send in 100,000 autonomous robots to form into geometric shapes at a football game, they will more less instantly spontaneously evolve into deadly killing machines bent on the destruction of humankind...
Re: (Score:1)
From past experience, simulators are great for replicating known scenarios (including a limited set of failure scenarios).
But real, physical things will fail in strange ways that are hard to predict or can be hard to simulate.
Eg, perhaps some of them have a slight tendency to lean to the left (uneven legs during manufacturing), some of them might be intermittently blind on the right side but only when turning right (maybe manufacturing machine leaves a dry joint when soldering the right sensor, not found du
They've invented Transformers! (Score:2)
Imagine a Beowulf cluster of these babies! (Sorry, I'm still on a retro kick.)
One thousand kilobots? (Score:3)
Is it one thousand kilobots (1000000) or one thousand killbots (1000) or one thousand kibibots (1024000)?
In the article they say it's 1024 bots, so whoever typed that title is probably smoking supercapacitors [slashdot.org].
Ah! Replicators!!! (Score:2)