?

Log in

No account? Create an account

Previous Entry | Next Entry

Cool Word of The Month: March, 2004

calad (Sindarin, noun) - light

Yeah, I think you know where this is going. :-D


Previous months' cool words:
February, 2004: su (Chinese, adjective/noun) - 1. flaky; 2. a baked good with a crisp or flaky consistency, such as a cookie
January, 2004: pizdarija (Croatian, noun; vulgar) - something messed-up, feeble, or ridiculous (definition provided by jereeza)
December, 2003: basherte (Hebrew, noun) - "apportioned one" (implication of predestined/ordained mate; courtesy of yahvah)
November, 2003: panmictic (English, adjective, "exhibiting random mating within a breeding population")
October, 2003: kreteno (Esperanto, slang noun, "idiot")
September, 2003: kawai (Japanese, adjective, "cute")
August, 2003: ser (Spanish, intransitive verb, "to be")
July, 2003: cordillera (Spanish, noun, "principal mountain system of a continent")
June, 2003: kallüsarayam (Tamil, noun, "illicit liquor")
May, 2003: hoh (Singlish, particle, "connective expression of expected agreement")
April, 2003: tmesis (English, noun, "separation of the parts of a compound word for humorous effect")
March, 2003: nerazreshimost (Russian, noun, "undecidability")


In other news: I'd like to write a survey article on autonomous nanorobotics.

Surprisingly, I have found much literature on nanotechnology, with rampant speculation on the role of nanorobotics. I've also found quite a bit on massively multi-agent systems as one aspect of the copious literature on autonomous robotics. I just haven't found much on both. Nanorobotics really has to be autonomous, if you think about it: communication takes quite a bit of energy and isn't well suited to organic or other insulation.

Any ideas?

--
Banazîr

Comments

scottharmon
Apr. 2nd, 2004 09:00 am (UTC)
Nanorobotics really has to be autonomous, if you think about it: communication takes quite a bit of energy and isn't well suited to organic or other insulation.

I would think that the comunication would be more subtle. For example, simple movements and behaviours could be a form of comunication to the collective group of nanobots. So, of course they probably won't be sending GB of maps back and forth, but simple protocols could be developed that allow them to communicate on some level (think ants or bees).
banazir
Apr. 3rd, 2004 07:52 am (UTC)
Autonomy, communication, and swarm intelligence
I would think that the comunication would be more subtle. For example, simple movements and behaviours could be a form of comunication to the collective group of nanobots.
So could pheromone trails and other "environment-borne" media, but it's hard to accurately simulate the dynamics or explain the emergent communication - i.e., capture the capacity, link it to channel coding, etc.
Ever read Cover and Thomas's Information Theory? hermes_imagod can tell you, it's good stuff. I keep trying to get Julie to look at it.

So, of course they probably won't be sending GB of maps back and forth, but simple protocols could be developed that allow them to communicate on some level (think ants or bees).
Dude.
Prey.
Crichton is lovenano.

See also this book whose link I posted to KDD yesterday. Actually, to look at the reviews, Sugarscape seems rather dinky. OTOH, there's a SourceForge project for it. I wonder if RoboSim could include a "massively multi-agent" scalability test bed. I think that would be a good research topic (both in terms of fundability and publishability).

Have you read Haipeng's survey work on ACO (Ant Colony Optimization, also called the AAAA track at GECCO)?
I still have Dorigo's tutorial from AAAI-2002, and there's ACO/PSO (Particle Swarm Optimization) work throughout the soft computing communities.

Personally, what I see is a lot of dynamic programming with ad hoc randomized dynamics tacked on, but there are a few "SA (simulated annealing)/GA-complete" ACO and PSO approaches (meaning that they have the same search power as an equivalent SA or GA). But if you really want something powerful, study the MCMC (Markov Chain Monte Carlo) books of Neal and Gilks, and take a course or read a book or two on stochastic processes. I can point you through the relevant particle filtering literature from UAI.

--
Banazir

Latest Month

December 2008
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

KSU Genetic and Evolutionary Computation (GEC) Lab

Teunciness

Breakfast

Science, Technology, Engineering, Math (STEM) Communities

Fresh Pages

Tags

Page Summary

Powered by LiveJournal.com
Designed by Naoto Kishi