« Conference oversold | Main | The liveblogging experiment »

Nanorobots

Professor Ari Requicha of the USC Lab for Molecular Robotics is speaking now about "Nanoassembly and Nanorobots" (abstract and bio here). He thinks the field of nanorobotics will need a new name soon, since it's gotten a lot of bad press in recent years -- thanks, for instance, to Michael Crichton's novel Prey -- and thus might sound scary here in Washington. There's some truth to that: a few months ago, the website of the federal government's National Nanotechnology Initiative pooh-poohed nanorobots, essentially describing them as mythical monsters. So a new name for nanorobotics research might be needed someday soon -- but not today, says Professor Requicha.

UPDATE: Prof. Requicha later said, in passing, that if he were to suggest an improved name for "nanorobots," it would be "sensor/actuators." Yup -- sounds innocuous enough.

The sort of robots Prof. Requicha is working would have to work in swarms, and would be very, very stupid -- they would follow a few very simple rules. But from simple rules, complex things could be built. They'd also have to be low-energy, and would have to communicate with one another chemically (as opposed to big mechanical robots today, which might for instance navigate and communicate with radio or sonar or vision). Prof. Requicha's robots "don't know where they are, they don't know if they're coming or going, they don't have a boss."

When asked afterwards about the potential abuse of these robots -- specifically, turning nanorobots into weapons -- Prof. Requicha responded, "Any powerful technology can go either way -- it depends on what you want to do with it.... Why would you bother to go through such an elaborate procedure, to build nanorobots, when you could just throw anthrax [at your enemy]."

I'll try to get some video up later of Prof. Requicha's constructive swarms, but I can't guarantee that it'll show up clearly.

UPDATE: Click the picture below for some streaming Windows video of an animation of Prof. Requicha's nanorobots at work. My understanding of what's happening in the video is this: the little red dots you see all over the screen (not to be confused with the bright red dot from Prof. Requicha's laser pointer) are supposed to be nanorobots. They are each given a very simple set of coded instructions, but are not told where to go. They just move about randomly, unless they happen to bump into a particular position where one of their simple pre-programmed rules is activated, in which case they execute that rule. (These are rules like, "if you bump into such-and-such a nanobot on so-and-so side, you should glom onto it.") The end result is the creation of a desired structure. (There was another animation where the nanobots spelled out the initials of Prof. Requicha's lab.)

At least that's my rudimentary understanding of it. I've taken out the sound from the video, to reduce bandwidth problems, but here's what the professor is saying at this point: "They [the nanorobots] first build the boundaries, you'll notice, and then they fill it up.... We can build arbitrary two-dimensional shapes. Whether we can do it in 3D -- I think we probably can, but it will be more difficult."

Requicha

October 22, 2004 in Research | Permalink

Comments

Oh boy. Perfessur Requicha is just the guy I want in the center of the nanorobotics research and ethics.

A technology can go either way. This guy didn't have anything to do with research on Africanized honey bees in South America in the 1960's and 1970's, did he?

Posted by: don | Oct 22, 2004 3:14:25 PM

"Just throwing Anthrax at an enemy" could be thought of as using a naturally-occurring self-replicating nanobot to attack someone.

Posted by: DensityDuck | Oct 22, 2004 10:05:27 PM

The comments to this entry are closed.