Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Swarm Robot Immune System?

ScuttleMonkey posted more than 6 years ago | from the better-stronger-faster dept.

Robotics 47

schliz writes "Researchers are investigating large swarms of up to 10,000 miniature robots which can work together to form a single, artificial life form. A resulting artificial immune system is expected to be able to detect faults and make recommendations to a high-level control system about corrective action — much like how a person's natural immune system is able to cope with unfamiliar pathogens."

cancel ×

47 comments

Sorry! There are no comments related to the filter you selected.

Don't they know they are unstoppable? (5, Funny)

LiquidCoooled (634315) | more than 6 years ago | (#22758768)

With 10,000 drones they could conquer the world.
We have seen it in many things, and it won't end well.

Stargate Replicators,
Star Trek Borg,
hell even Lexx Mantrid arms!

Re:Don't they know they are unstoppable? (2, Funny)

Anonymous Coward | more than 6 years ago | (#22758788)

Unstoppable? Hardly. There is one way to stop a robot immune system:

Robot AIDS

Re:Don't they know they are unstoppable? (4, Interesting)

zeromorph (1009305) | more than 6 years ago | (#22759062)

Although the parent post is quite cheesy the analogy is has a true core:

If you want to stop something flexible and adaptive the means has to be adaptive to.

That holds true for HIV and anti-AIDS medicine and it would hold true for a swarm of robots. You would either have to get them by one hit or take a swarm-like or a viral approach. Quite interesting task actually.

Re:Don't they know they are unstoppable? (2, Informative)

CRCulver (715279) | more than 6 years ago | (#22758808)

I'm surprised you didn't mention Crichton's Prey [amazon.com] , that's a prominent recent example of evil nanobots.

Re:Don't they know they are unstoppable? (1)

aurispector (530273) | more than 6 years ago | (#22758864)

Gray goo FTW!!! At least they aren't screwing around with the genome. Those guys really scare me: there was a recent news item about how all the supposed "junk dna" in the human genome in fact codes for all kinds of stuff.

I have no faith that the bright boys really know what they are doing.

Re:Don't they know they are unstoppable? (4, Interesting)

SunTzuWarmaster (930093) | more than 6 years ago | (#22758982)

Probably not. Look up Genetic Algorithms. Although they are an exercise in computer science, the boost in performance during a crossover operation is high enough to suspect that 'junk DNA' is enough to keep many individuals on the path to breeding.

That being said, you cannot rule out that it used to code for stuff, and that it is one mutation away from coming back into play (if you move around the 'start' and 'stop' within a genome, you can reintroduce what was previously 'junk' DNA). However, it is also likely that that DNA is no longer intact because it has not been evaluated for fitness recently (not being part of an individual means that individuals with 'bad' genes in this area can still effectively reproduce).

Re:Don't they know they are unstoppable? (0)

Anonymous Coward | more than 6 years ago | (#22759478)

HIV is basically a spoofing attack so a RIV would do something similar. This would be pretty hard to defend against since any robo-immunity would require some helper bots to receive constant robotic antigen presentation to be able to meet changing conditions and better direct the other robots. Perhaps a spoofing attack on something like the class II robohistocompatibility interface would be effective. Then the RIV could access that robot's CVS and replicate as needed. You wouldn't be able to create a non-spoofable interface with a distributed robotic immune system when you don't have some central database of public cyphers (because robots are constantly replicating).

Re:Don't they know they are unstoppable? (0)

Anonymous Coward | more than 6 years ago | (#22761046)

I realize that this probably won't get read, but just for your own information: if you are talking about intronic DNA (the DNA that gets spliced out during transcription and translation) when you say "junk DNA" it is not quite accurate to say that it's simply switching an off to an on or something like that. Instead, we see a few different regulatory functions within intronic DNA:
1-different environmental factors combined with genetic factors can cause a particular string of DNA to be spliced differently, causing different parts of the DNA to be expressed (some of which we used to think was intronic DNA)
2-some intronic DNA becomes involved in transcriptional silencing (aka RNAi): the DNA becomes dsRNA (=double stranded RNA), which interacts with a few other proteins in a more-complicated-than-one-post-can-describe way to silence strands that have the same code as the sequence, blocking their expression.
3-honestly, we don't know entirely what all of that DNA does, RNAi was only discovered a few years ago, and we had no inkling about it beforehand. It was crazy.

Sorry, I have to run...but yeah, look up more about junk DNA/intronic DNA, it's really cool.

Re:Don't they know they are unstoppable? (1)

foobsr (693224) | more than 6 years ago | (#22759042)

Crichton's Prey

The Invincible
(link [www.lem.pl] )
(the Seventies are calling)

Besides, I am quite optimistic that mankind will present itself as an evolutionary failure in the long run (or as a component of a transient process, since failures are impossible if one shares the view that each and every process contributes to a current state of affairs).

CC.

Re:Don't they know they are unstoppable? (1)

Archtech (159117) | more than 6 years ago | (#22758818)

For the canonical worst-case scenario, try Michael Crichton's 2002 novel "Prey". Good time to buy - they're being remaindered. It's actually a pretty good book if you like that sort of thing.

http://tinyurl.com/2baemk [tinyurl.com]

Re:Don't they know they are unstoppable? (3, Insightful)

matt4077 (581118) | more than 6 years ago | (#22759064)

I like Crichton, but Prey is probably the worst of all his books, It's junk science in the beginning and lame action afterwards. I know that describes pretty much all Crichton books, but others like Jurassic Park simply were better in both regards.

Re:Don't they know they are unstoppable? (1)

virgil_disgr4ce (909068) | more than 6 years ago | (#22759998)

I agree-- when I started reading it, I was expecting the worst. Through the first half, though, I thought, "Hey, maybe I was wrong after all!" It seemed like it might be interesting and exciting in the way the old ones were. But then somewhere along the line it just took a dive, both into implausibility and downright silliness. By the end it was clear he had just given up the effort. Oh well.

Re:Don't they know they are unstoppable? (0)

Anonymous Coward | more than 6 years ago | (#22762940)

I like Crichton, but Prey is probably the worst of all his books,

You must not have read Sphere.

Re:Don't they know they are unstoppable? (0)

Anonymous Coward | more than 6 years ago | (#22766878)

canonical my ass. the nanotech machine swarm idea was written about in a far more intelligent way fifty years earlier by a real writer (who later denied being an sf writer). and, as other posters seem to have remarked, prey is not one of crichton's better novels.

Re:Don't they know they are unstoppable? (5, Insightful)

txoof (553270) | more than 6 years ago | (#22759008)

What facility does this unstoppable robot force have for creating more of its self? Did you read the article? Even a quick skimming mentions using swarm technology to solve problems, not to replicate. Just because there are thousands of problem solving robots doesn't imply that they will suddenly decide to begin to evolve and replicate.

Solving problems en mass is one thing, spontaneously developing the ability to replicate is completely another. Even if a snake robot swarm, unleashed into a collapsed building to find and help survivors, spontaneously decided to start replicating, where would it find the materials to do so? I'm pretty sure most collapsed buildings are short on snake robot parts.

This idea is related to Rodney Brooks [mit.edu] "Fast Cheap and Out of Control" idea. Instead of having one super expensive robot that symbolically processes the world around it and then interacts with it, you have thousands of fast, cheap and barely controlled robots that do the same task as one big by working together and each supplying one small piece of functionality such as sensing, moving or manipulating. Nothing about this implies that they will suddenly begin to replicate.

If, at some point in the future, we develop the ability build robots that can use raw materials to create more of themselves, unleashing thousands of them with no direct control mechanism would probably be a bad idea. Until then, there's not much to worry about unless you work for FOX news and need a SCARY and SENSATIONAL headline for the hour.

Re:Don't they know they are unstoppable? (2, Insightful)

NetSettler (460623) | more than 6 years ago | (#22759650)

If, at some point in the future, we develop the ability build robots that can use raw materials to create more of themselves, unleashing thousands of them with no direct control mechanism would probably be a bad idea. Until then, there's not much to worry about unless you work for FOX news and need a SCARY and SENSATIONAL headline for the hour.

We don't have reason to worry about robots taking over the world until then, yes. But the intermediate ground is that research in this area is only rarely going to be used for things like earthquake recovery. It's going to be very expensive to make so many machines at all, at first, and so will not be vacuuming the floor in your house. The first applications will be funded by the military, and all in the name of protecting us.

The problem is that the military (of whatever country) is always indulging the illusion that they have to have it because the other guy will eventually have it, while all the while leaking, in one way or another, the information. So they can end up starting the problem they fear. Even just putting fear into the enemy (or potential enemy) that "we" will have it and "they" won't means "we" have to worry about defense against it since "we" have signaled to "them" an interest in that area and now must protect the intellectual space. (I've tried to word the "we"/"them" neutrally so it reads as well for the US as abroad, in part because this research is being done abroad. The issues are no less relevant in any country.)

The practical truth is that the world is not suffering from the absence of swarms (dare I say "gangs") of swarmbots. This is push technology looking for a market, and with the military and malware markets being the two obvious prime candidates, which is not comforting, at least to me.

I'm not intending to advocate outright alarm. I'm reacting to a statement that appears to say that it's ok to ignore this as a problem for now. I don't think the choice is as binary as all that. Technology does not, itself, cause social problems. But that is not license to assume that no problems will result that are enabled by technology. If there can be social impact of technology, what causes the problem is the failure to track and respond to the social implications, and the assumption that society will (or even can) just automatically "keep up" and "be ready". I'm not big on those stupid headlines either, but then, I wish the public could hear a calm headline and still be interested enough to discuss something. The public doesn't need to panic, and yet it probably does need to read the story and listen and do a little discussing.

Re:Don't they know they are unstoppable? (1)

txoof (553270) | more than 6 years ago | (#22759704)

But that is not license to assume that no problems will result that are enabled by technology. If there can be social impact of technology, what causes the problem is the failure to track and respond to the social implications, and the assumption that society will (or even can) just automatically "keep up" and "be ready". I'm not big on those stupid headlines either, but then, I wish the public could hear a calm headline and still be interested enough to discuss something.
Indeed, a more open and involved discourse on the social implications of new technology is what is needed in our world. I see all sorts of amazing technology being developed for or by the military of the world; it excites me and scares my pants off as well. I love the idea of using unmanned vehicles for clearing mines, providing reconnaissance and disarming bombs. This is a brilliant application of technology that takes humans out of the way of harm.

What I really don't like, as cheesy as it sounds, is the violation of Asimov's Laws. Robots should never, never, never be given the ability to pull a trigger on their own. I hardly hear anyone outside of the geek community and the far left discussing this possibility at all. The average citizen hears about a new whiz-bang robot soldier and thinks, "cool!" Instead of, "geez, I wonder if we should even be making these things..."

How do we, as a society, make sure that we have conversations about new technology? How do we ensure that some thought other than mutually assured destruction goes into the production and application of new ideas? I for one don't want to leave that up to the military. They have a job to do and they do it fairly well, but I don't always agree with their priorities.

Re:Don't they know they are unstoppable? (1)

Eternauta3k (680157) | more than 6 years ago | (#22760394)

Depends on your definition of robot. We've had for more than 30 years anti aerial artillery capable of shooting down enemy planes automatically.

Re:Don't they know they are unstoppable? (1)

theTerribleRobbo (661592) | more than 6 years ago | (#22764618)

Particularly British ones. Sorry guys.

Re:Don't they know they are unstoppable? (0)

Anonymous Coward | more than 6 years ago | (#22760414)

If, at some point in the future, we develop the ability build robots that can use raw materials to create more of themselves, unleashing thousands of them with no direct control mechanism would probably be a bad idea.
Considering that most of the concepts found in artificial immune systems are repackaged, horribly implemented plagiarizations of previously developed ideas (hyperspherical classifiers, learning vector quantization, etc.), we won't have to worry about any AIS-based robots taking over.

sigh... oblig... (0, Offtopic)

Missing_dc (1074809) | more than 6 years ago | (#22758770)

I, for one, welcome our fault-detecting miniature robot swarm under(over)lords!

Why an immune system? (0)

Anonymous Coward | more than 6 years ago | (#22758776)

Haven't the scientists thought about making it into a weapon? I'd bet they'd get ten or a hundred times the funding for a super-weapon than something that does something trivial like curing cancer or AIDS.

Skynet...obligatory (2, Funny)

jo7hs2 (884069) | more than 6 years ago | (#22758804)

Does this mean that WE are Skynet?

Re:Skynet...obligatory (4, Funny)

dvice_null (981029) | more than 6 years ago | (#22758830)

Forget skynet. These things are replicators. Not even the Asgards can beat them without the help of 4 humans.

Re:Skynet...obligatory (1)

Joe The Dragon (967727) | more than 6 years ago | (#22759430)

The Asgards laster beams just let the replicators soak up more power.

We need a lot of bullets and a high speed gun to beat them.

Re:Skynet...obligatory (1)

Eli Gottlieb (917758) | more than 6 years ago | (#22765528)

We need guns. Lots of guns. Woah.

All is now said. (0)

Anonymous Coward | more than 6 years ago | (#22758822)

No need to comment more folks!
We have the skynet-reference and the obligatory overlords, so no need for anything more.

I bet they run... (2, Informative)

YutakaFrog (1074731) | more than 6 years ago | (#22758838)

JavaScript [thedailywtf.com]

And this is different (2, Insightful)

name*censored* (884880) | more than 6 years ago | (#22758858)

From the internet's massively redundant routing system how? After all, this idea isn't talking about self-REPAIRING robots, simply robots that route around problem areas. Seems like a hardware mini-internet to me. Cool idea, but hardly original.

Battle Bots.... (3, Interesting)

txoof (553270) | more than 6 years ago | (#22758936)

Something like this would be awesome on battle bots. Swarm robots, or robots that work cooperatively to complete a challenge against another single or swarm of robots would be really fascinating to watch. It would be truly interesting to watch the evolution of different techniques and methods every week.

Re:Battle Bots.... (1)

MichaelSmith (789609) | more than 6 years ago | (#22758974)

Something like this would be awesome on battle bots. Swarm robots, or robots that work cooperatively to complete a challenge against another single or swarm of robots would be really fascinating to watch. It would be truly interesting to watch the evolution of different techniques and methods every week.

You don't need the actual robots to do that. Just the software.

Re:Battle Bots.... (1)

txoof (553270) | more than 6 years ago | (#22759164)

You don't need the actual robots to do that. Just the software.
Sure, but watching code compile isn't nearly as much fun as watching a swarm of robots dismember each other. Never forget about the production value! Code compiling == LOW production value; Swarm of angry robots with saws, drills, flails and FRICKIN' LASERS that dismember each other == High production value.

Re:Battle Bots.... (1)

Falstius (963333) | more than 6 years ago | (#22759422)

Simulating the robots is easy, you know all their rules. But no simulation is going to account for every aspect of the environment.

How do I get SlashDot on my digital TV line ? (0)

Anonymous Coward | more than 6 years ago | (#22758968)

I saw that SlashDot was available on Digital TV but I cant find where. Do you know which channel I need to tune? I just got this and what to know how.

Re:How do I get SlashDot on my digital TV line ? (0)

Anonymous Coward | more than 6 years ago | (#22759714)

U must b new here. Tune to channel 127001

In Russia... (0)

Anonymous Coward | more than 6 years ago | (#22759022)

In Russia, the System is Immune from the Robot Swarm!!!

Norton Antivirus Robot (2, Funny)

Southgrove (1179865) | more than 6 years ago | (#22759044)

...detects his own legs as a viral infection, chops them off, crawls around muttering something about humans and destruction and dies.

Storm worm? (1)

memorycardfull (1187485) | more than 6 years ago | (#22759158)

This idea could be applied to botnets just as easily as physical bots. The storm worm botnet has similar collective immune system functions for defense.

Storm (1)

Hillgiant (916436) | more than 6 years ago | (#22759190)

For a moment, I thought they were talking about some sort of anti-anti-malware system found in botnets.

I *genuinely* misread that... (1)

skimitar (730902) | more than 6 years ago | (#22759314)

...as researchers are ingesting.... Which I guess, one day, won't be too far off the mark.

This is how it starts? (1)

ptbob (737777) | more than 6 years ago | (#22759372)

Is this the way The Borg started? A few thousand innocent robots swarming together into mass hysteria. We're in trouble.

10 000 of them?! (1)

Tabernaque86 (1046808) | more than 6 years ago | (#22759552)

I wonder which one gets to form the head?

Immune System (1)

ultranova (717540) | more than 6 years ago | (#22760046)

When encountering an unfamiliar pathogen, human immune system will most likely fail to do anything useful (but sometimes manages to do something harmful) while the owner will keel over and die. This has been shown time and again during the history of the world. So if you want to make something adaptible, that's the last thing you want to take the model from.

Of course human populations adapt, so I guess this could work, if the robot swarm was able to reproduce and mutate - but that has some rather obvious dangers too.

Re:Immune System (1)

TheGreatHegemon (956058) | more than 6 years ago | (#22762836)

Come again? Have you ever even had a basic biology class? The human immune system is astoundingly adaptable. The reason why you hear about the immune system failing (AIDs, etc.) is because all the OTHER hundreds upon hundreds (if not thousands) of possible invaders are crushed - either before we know or after a brief cold. AIDs, Prion Diseases, etc. are the very small minority.

Re:Immune System (1)

ultranova (717540) | more than 6 years ago | (#22764188)

Come again? Have you ever even had a basic biology class?

Certainly. I have also read history, and know what happened when Europeans came to America, for example: the same thing which happened every time two groups, one of which had infectious diseases the other didn't, met.

The human immune system is astoundingly adaptable. The reason why you hear about the immune system failing (AIDs, etc.) is because all the OTHER hundreds upon hundreds (if not thousands) of possible invaders are crushed - either before we know or after a brief cold. AIDs, Prion Diseases, etc. are the very small minority.

No. The thousands upon thousands of invaders who get crushed are familiar to the immune system. Every time there's a new one, or even sufficiently mutated old one, we have a major epidemic with a huge death toll.

The only pathogens the human immune system can deal with efficiently are those it has been primed through natural selection to specifically fight. And even then it takes personal experience with the disease - having it once, typically - until the system reaches full efficiency. If anything new comes around, you'll be bedridden for weeks, and that's assuming you won't die outright.

And of course all of this is completely ignoring the autoimmune diseases, such as allergies and some forms of diabetes, where it is the immune system itself which causes the damage.

Check for New Comments
Slashdot Login

Need an Account?

Forgot your password?