Swarm Robot Immune System?

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."

Comments

Don't they know they are unstoppable?

[LiquidCoooled]

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?

[Anonymous Coward]

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

Robot AIDS

Re:Don't they know they are unstoppable?

[CRCulver]

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

Re:Don't they know they are unstoppable?

[SunTzuWarmaster]

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?

[txoof]

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 "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?

[zeromorph]

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?

[matt4077]

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?

[NetSettler]

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.

Skynet...obligatory

[jo7hs2]

Does this mean that WE are Skynet?

Re:Skynet...obligatory

[dvice_null]

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

I bet they run...

[YutakaFrog]

JavaScript

And this is different

[name*censored*]

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....

[txoof]

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.

Norton Antivirus Robot

[Southgrove]

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