Slashdot Mirror
SwarmOS Demonstrated at Idea Festival
PacoCheezdom writes "Intelligent Life has short summary of a demonstration by MIT professor James McLurkin of his new group-minded robots, which run an operating system called 'Swarm OS'. The robots are able to work together as a group not by communicating with all members of the group at once, but by talking only to their neighbors, and model other similar behaviors performed by bees and ants. "
I, for one, welcome the new swarming overlords nearest to me, so that they might welcome the rest.
@AlexSheive
I worked in robotics for 3 years and there was a big fad of cooperative robotics. Now, closely related is this swarm stuff. But theoretically it is the same as having a robot with many parts (i.e. higher dimensional phase space). I never saw any real applications.
You can read more about this research and see some videos of the robots in action here.
Thank you for the welcome! I, for one, welcome the swarming robotic overlord bits below me. Hopefully we can figure out how to build a petrified, robotic Natalie Portman and cover her in naked robotic grits. (Is this possible, considering we can only talk to our neighbors?)
Legalize it.
I saw Mr. McLurkin give his presentation here in Ottawa. Fascinating stuff. Each component of the swarm is very dumb, with very little storage. If you want to store a location for future reference, it's very easy; park a robot there.
All the robots have a sound system, though; the first thing Mr. McLurkin did during his presentation was to have a single robot request that 6 other robots follow it, and the swarm picked and allocated 6 robots, and they all went off in a chain, singing "Hi-ho, hi-ho, it's off to work we go".
Check out James McLurkin's website for some presentations and videos:
http://people.csail.mit.edu/jamesm/
...of an obligatorily /.-ted article here.
I feel a beowulf cluster joke coming on...
Boids was a program written to try to simulate the flocking behavior of birds. It was written by Craig Reynolds
Reynolds gave his boids 3 rules:
1 Don't crowd too close to other boids
2 Try to go the same direction as other boids near you
3 Try to be in the average position of your local neighbors.
With just those three simple rules, the boids arranged themselves in a flock. Much to Reynolds surprise, without any more rules than that, the flock exhibited other emergent behavior, such as a flock that split up to go around an obstacle would rejoin.
More at: http://www.red3d.com/cwr/boids/
Here is the Google cache if anyone is interested.
Insert Generic Sig Here:
Practical application: self-laying mines. Think how annoying it would be to clear a path and then overnight see the 95% of the mines you missed on day one redeployed in near-randomness across your path back.
(Yes I have MOD points today...it's just more fun to talk.)
Bees and ants seem to be a good argument that it might be a good return on investment. So do search parties when looking for lost hikers.
Robot-1: I heard from Robot-2 that Robot-3 got promoted because she slept with the boss after the Christmas party last year.
Robot-4: I knew something was going on. Robot-3 doesn't even have opposing digits, how can she be qualified for the ball in bucket tests?
With all the advancements and buzz about smarter-than-man AI, and now they're excited about robots as smart as BEES AND ANTS.
10 FILL MUG WITH COFFEE
20 DRINK COFFEE
30 GOTO 10
Simulated robotic orgies.
Eternity is a time bomb.
Some are complaining that they can't get to the server, so here is the text:
...
Special to INTELLIGENT DAILY LIFE
Surrounded by buzzing robots that end the session by performing in an orchestra, James McLurkin, a PhD student at MIT Computer Science and Artificial Intelligence Laboratory, talks about distributed robotics and swarm behavior to a packed house. His work has its roots in "swarm intelligence"--the study of collective behavior in decentralised, self-organised systems. Think of ant colonies, bird flocking, animal herding, fish schooling, and many other examples in nature.
During the last few hundred million years, nature has perfected such interactions. Now, scientists such as McLurkin want to get a better understanding of how these biological processes work and apply this knowledge to programming robots for doing complex tasks in groups. Perhaps, this is the ultimate interpretation of the Wisdom of Crowds thesis: individuals don't have to be smart to produce very smart group outcomes. Did somebody mention Wikipedia?
Early on, McLurkin pulls up a slide of Isaac Asimov's famed three laws of robotics, intended to forestall a robot revolt against humanity. "Well, robots don't know how to read, so those laws are not particularly useful", he smiles. Robots are not even smart enough to travel from the stage to the audience: they would get trapped in wires or collapse to the floor. For all the talk about robotics, today an average squirrel can still do more than any robot, he says.
He points to a number of philosophical, not just engineering problems, in his field. Problem number one is that we don't know what intelligence is, nor how to define it. Should we subject the robots to some upgraded version of the Turing test (which says that if a judge can't tell whether he is talking to a machine or a person, the machine passes the intelligence test)?
Can intelligence emerge from interactions of unintelligent components? That is a second philosophical question. As we are all built from molecules, continues McLurkin, either intelligence is something that results from interactions, or molecules are intelligent.
The third and final question is whether an intellect needs a body. Can a brain in a vat understand and experience the world without anything to relate to? Can we build such an intellect?
That slide with the three philosophical questions is subtitled "things that make you go "hmmm", and one can hear half of the audience "hmming".
Having finished with the philosophy, McLurkin gives a brief overview of earlier efforts to mass-build robots, presenting quite a few models, from iRobot Roomba to Honda Asimo to iRobot Packbot, all of them having different looks and different functionality. And, of course, NASA's successful launch of two robots on to Mars.
Quite naturally, he makes a transition to his own work. He has 112 robots in his arsenal and he is trying hard to make them work together. In his view, robots are best at jobs that are dangerous, dirty, or dull: "What if we sent 20 robots to work in hot spots around the world? What if we sent 200 robots to look for surivors after an earthquake? What if sent 2,000 robots to explore Mars?".
It's this last question he wants to address with his on-stage demonstration. McLurkin turns to a few dozen robots that he has on stage (he controls them with a remote). As a starter, he asks the robots to form a line; surely enough, they do. Next, he orders the robots to spread out. They do this too. The demonstration proceeds quite smoothly.
One thing that the robots don't know yet is how to define boundaries of the network, so they often spread out from the center and then get disconnected. The robots can communicate via one another (they know the neighbors, but don't know about everybody else) but
In his second dispatch from the Idea Festival in Louisville, Evgeny Morozov watches a podium-full of robots buzz around like bees, ask each other questions, find an orange, leave the room, form an orchestra, and prepare one day to save your life
weirdest thing I ever saw: scientology advertising on slashdot.
Swarm of UAVs for surveillance of hostile (and friendly?) countries. UAVs work together to accomplish goals such as "make sure there is a flyover of areas X Y and Z every 10 minutes", "keep a unit no less than 5 minutes away from this location", "keep 20 units in the airspace, but make sure each unit charges to at least 40% at all times."
Professor James McLurkin now goes by the designation "1 of 12".
It must have been something you assimilated. . . .
I for one welcome the new swarming overlords nearest...oh - I've just received word from the swarm that someone already posted this. Ok how about:
In Soviet Russia, bots swarm...oh - that too? Ok, how about just a simple "Profit?"
I think this swarm thing will take some getting used to
From TFA : "...James McLurkin, a PhD student at MIT Computer Science and Artificial Intelligence Laboratory..."
He's a student, not a professor. Way to read the article, Mr. submitter.
An application of what, specifically? Machines that talk to each other and not to the whole group and do something useful? You mean, like Bittorrent? :)
Seriously, though, this is some very cool research; the robots talk to each other via infrared, which is why they can only talk to their neighbors. But, with the infrared setup they're using, they can estimate direction and distance to each of their neighbors. You COULD do this with a bunch of robots talking bluetooth with GPS receivers, but it would be insanely expensive by comparison. These guys would be dirt cheap if mass produced. Dirt cheap means you don't care if you loose a few, which makes them excellent options in harsh environments.
A lot of the research in this area right now is in algorithms. Designing an algorithm to run distributed over a group of small, dumb, physical devices, where individual devices might suddenly disappear (batteries die, fall down a hole, consumed by fire, eaten by ewoks, etc...) is quite difficult.
If you're looking for a practical application; thousands of bug sized robots which scour a collapsed building for survivors, and direct rescue efforts. If you loose a few, who cares? They're cheap! Or how about thousands of small rovers which explore the surface of an alien planet? If you send a single rover to Mars, you're putting all your eggs in one basket; if something goes wrong with that rover, you're whole mission fails. A collection of cheaper robots which can work together dramatically looses your odds of failure, since the failure of 10% or 20% of the swarm would be immaterial.
Comment removed based on user account deletion
I, for one, welcome the swarming robotic overlord bits below me as well.
Mmm... Grits...
Swarming and flocking behavior also inspired a freeware game called Swarm Racer, in which you get to control a swarm of micro-racing robots. For Windows and Mac OS X.
Imagine a Beow... wait a second...
Cars driving themselves.
Can you imagine a bee or wolf cluster of these?
I'll be here all week.
"They said I probly shouldn't fly with just one eye," "I am Bender. Please insert girder."
I'm sitting idly waiting for my parent post to welcome me so that I can pass it on to my children posts.
How are humans and bees/ants/swarms different?
When you put lots of humans together, they get dumber.
My neighbors don't like me. :(
There goes my view of the world.
I know I already replied, but I was just reminded of a quote. I forget whether it was Babbage or von Neumann or whoever, but he had just finished giving a talk about the computer. During the question and answer period a woman raises her hand to ask, "That's all very nice, but of what use is it?" To which he replies, "Madame, of what use is a newborn child?"
I'd be glad if someone could tell me who said this or if it's apocryphal or whatever. I looked on google a bit with no luck.
Give me Classic Slashdot or give me death!
practical applicationS: Airplans flying and not crashing into one another. Same for cars.
More practical. How about Earthmoving equipment or coal mining.
Some exotic ideas. Military robots that gather intelligence. You
drop thousands of these on the enemy's side and they look out to see what is going on and report back via "the grape vine". There would be tens of tousands of communications paths, far to many to jam. They also watch out for each other and communicate warnings like "hide, someone is coming." Sensor could be very primitive, perhaps just a microphone or a cellphone-like camera, but by working together they can use triangulation to locate moving targets.
They don't have to be robots. What about a self configuring network? Each node only sees a few other nodes but they all talk about what they've seen and the word gets around that there is a printer on the second floor available for anyone who is a member of the graphic arts department to use.
Would you say that you "Live for the Swarm?"
Can you be Even More Awesome?!
No, it isn't.
Swarm intelligence relies on emergence that arises from many simple agents that interact locally with each other (i.e. without a master controller), using minimal rules. These are the keypoints of this field: there isn't a single point of failure, you can ensure degradation of service gracefully, you can even perform self-repair, etc. It allows to solve large problems without having to implement a complex monolithic system.
Naturally, the difficulties then lie in defining the right rules for the swarm of agents to generate the expected macroscopic group behaviour.
I know of at least one shipping company (in Switzerland IIRC) that uses swarm intelligence techniques to give a good solution to the TSP in reasonable time, and uses that to schedule the plan for its trucks daily. More applications are being developed, including small robots to inspect parts in inaccessible locations (e.g. airplane), etc.
theefer