[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
formal models
From: |
cal |
Subject: |
formal models |
Date: |
Mon, 19 May 97 12:38:40 PDT |
mostly for vladimir -
'affine spaces' is the correct term to use in english
(if the space has an actual origin, then it is a vector space or linear space,
but if there is no distinguished origin, then it is an affine space)
can you describe the qualities you are looking for in your model?
it seems as though they are (at least):
location
motion
interaction
but i don't think that behavior is easily represented as a vector in such a
space,
unless it is very high-dimensional
one of the difficulties will be that the agents themselves contain state
information,
and either that gets represented in the n-space
(which means that all the agents have identical sets of parameters),
or else the n-space is only to represent locations,
and something else representes the states of the agents in the space
it seems to me that the kind of space you want is a variable-dimension space,
with parts in some dimensionalities and parts in others
(representing the fact that the important criteria for agent behavior is not
homogeneous across the entire space) -
this also allows locally affine spaces to be used,
and if care is taken with the transitions,
the entire model can be made appropriately formal,
while still having a structure that reflects the way it would be implemented in
swarm
keep working on the qualitites you want
more later,
cal
Dr. Christopher Landauer
National Systems Group, The Aerospace Corporation
The Hallmark Building, Suite 187
13873 Park Center Road, Herndon, Virginia 20171
e-mail: address@hidden
Phone: (703) 318-1666, FAX: (703) 318-5409
==================================
Swarm-Modelling is for discussion of Simulation and Modelling techniques
esp. using Swarm. For list administration needs (esp. [un]subscribing),
please send a message to <address@hidden> with "help" in the
body of the message.
==================================