[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Density Graph, How?
From: |
Dr. Paul Box |
Subject: |
Re: Density Graph, How? |
Date: |
Thu, 2 Oct 1997 09:23:36 -0600 |
Actually, this leads into another question I've been pondering... I am
cross-posting this message to swarm-modelling, as that is where this part of
the
discussion really belongs. I would suggest that any follow-ups be posted there
as well.
How do you calculate density of your agents (assuming that you are using a 2d
space or something of the sort)? There is the standard method of dividing your
space up into quadrats or whatever unit space you're using, and simply counting
the number of agents within the area at each step. That is probably the most
tried and true method. However, I have been playing around with a c library
written by Dave Watson, at the University of Western Australia, that has an
extremely fast (and elegant) way of determining the nearest neighbors of points
on a surface. I was using the code to have an agent query it's immediate
neighbors for certain properties at each step of the simulation.
In the process of determining nearest neigbors, the code also calculates
thiessen polygons around the points. Since each thiessen polygon is an area
with exactly one point (agent) inside it, then the area of the thiessen polygon
is a measure of that agent's personal space. I was thinking of using one over
the area (area^-1) as a quick measure of that agent's density packing, which
would then become a property of that agent, which would affect it's behavior at
the next time step.
Is anyone out there aware of precedents of this in the existing literature? I
haven't found any, but that may mean I'm simply looking in the wrong places.
Cheers
>
> Hi,
> Thank you for all who replied my email.
> However, I was more interested in drawing the density graph for
> each generation of SWARM application....
>
> -Jae
>
>
==================================
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.
==================================