proposal: perception

From: Michael Brenner (brenner@informatik.uni-freiburg.de)
Date: Mon 15 Nov 2004 - 13:42:05 GMT


The current way percepetion is modeled is unsatisfying in a number of
respects:

1. In reality, if people have clear view, they can see much farther than 10m
2. However, they cannot see into houses even if they stand directly in
front of it
3. Since vision is calculated according to the reference points of
objects, there are some unnatural effects (described nicely in
Morimoto's manual, section 3.5). One very inconvenient aspect is that
agents may block each other without seeing each other!
4. Agents cannot see what other agents are doing even if it is something
as visible as extinguishing a building!

Here is a rough draft on how these problems could be overcome:

1. A possible solution could be the computation of specific visibility
areas for each position on the map by some simplified for of ray
tracing. Details are open to discussion, but having the GIS information
it is not too difficult to calculate which objects are visible from a
certain point on the map. If we had the height of buildings (or the
number of floors) we could even describe that one can see a big building
behind a small one, but not vice-versa.

A nice additional feature could be that visibility goes down in areas
where there is a lot of smoke.

2. To find buried civilians, agents should go into the buildings.
Finding the civilians should also *take time*, depending on the size of
the building, its degree of destruction, its fieryness, the number of
people hidden there, maybe the floor on which they were before the
building collapsed etc.

3. This bug would hopefully be fixed by 1.

4. It is difficult to tell what exactly an agent would see when
observing another one acting. For instance, if A observes B
extinguishing a fire, A would probably sense which building B is aiming
at, but not the water quantity. So we must be careful modeling such
extended sensing. Still, it would make coordination among agents in the
same area more natural.

What do you say?

        ResQ Freiburg (Michael)

-- 
Michael Brenner 		Phone: (+49) (761) 203-8226
Institute of Computer Science	Fax:   (+49) (761) 203-8222
Albert-Ludwigs-University	mailto:brenner@informatik.uni-freiburg.de
Am Flughafen 17, Geb. 52	http://www.informatik.uni-freiburg.de/~brenner
D-79110 Freiburg, Germany



This archive was generated by hypermail 2.1.3 : Mon 15 Nov 2004 - 13:58:38 GMT