Re: Real Space

[Darren Schreiber]

In my political party formation model, the agents are positioned in a 
continuous issue space.  Their preferences on n issues are chosen 
randomly from a uniform distribution.  They look for their closest 
neighbor in the issue space and form political coalitions with that 
neighbor.  I illustrate this for two dimensions at a time during runs 
using the canvas and graphs toolkit.  You can see the resulting 2d 
animations at http://www.bol.ucla.edu/~dschreib.

Similarly, my toy cocktail party model has agents who are able to 
only see a certain distance and look for people they are attracted to 
within their vision.  That model is at 
http://zia.hss.cmu.edu/econ/homework00/2/cocktail.html.  This also 
uses the graph and canvas.

The cocktail code is very simple and gives a basic idea of how to use 
some of the canvas and graph libraries. I coded it in only a couple 
of hours, so it has some kludges, but you can get the idea.

So, the political party model is an instance of looking for your 
k-nearest neighbors (where k=1).  And, the cocktail party model is 
instance of looking for neighbors within radius x.  While I 
illustrate that with the canvas and graph, the search of the space is 
done by parsing through the lists of agents and is independent of the 
display issues.

Let me know if you have any questions.

        Darren



At 12:24 PM -0400 4/11/01, William M. Rand wrote:
> 	I've been trying to read through the archives and so far seem 
> to have not
> had any luck.   Basically I want to implement a series of agents in a real
> (contiguous) space.  I noticed that many of the swarm simulations that I
> have seen that look like they are actually in a real space are actually just
> in a highly discretized space.  Is there anyone who has actually implemented
> a truly real space?  It would be nice to be able to have real coordinates
> for an agent as well as being able to find "neighbors within radius x" or
> "k-nearest neighbors"
> -Bill
>
> Bill Rand    1855 Shirley Lane #A8   Ann Arbor, MI 48105    734-623-1970
> address@hidden              http://www-personal.engin.umich.edu/~wrand/
>   "I must not fear. Fear is the mind-killer. Fear is the little-death
>  that brings total obliteration. I will face my fear. I will permit it
>  to pass over and through me."-Litany against Fear, Dune, Frank Herbert
>  \=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/\=-+-=/
>
>
>                   ==================================
>    Swarm-Support is for discussion of the technical details of the day
>    to day usage of Swarm.  For list administration needs (esp.
>    [un]subscribing), please send a message to <address@hidden>
>    with "help" in the body of the message.

--

_____________________________________________

                 Darren Schreiber
                  Attorney at Law
                 Graduate Student
             Political Science, UCLA
                address@hidden
        http://www.bol.ucla.edu/~dschreib

                 ==================================
  Swarm-Support is for discussion of the technical details of the day
  to day usage of Swarm.  For list administration needs (esp.
  [un]subscribing), please send a message to <address@hidden>
  with "help" in the body of the message.