Re: [robocup-rescue-s] 2007 Competition Comments (until now)

Dear all,
 
In line with ongoing discussions on maps used, I would like to point out
that the TC/OC have recognised the problem in this area and discussed it
with participants a few days ago. It has then been decided that next
year's competition will require maps as entry material for the
competition. We are quite keen on going ahead with this and any comments
or suggestions on how to best do this will be greatly appreciated.

It could also be useful if we can have a repository of maps on
sourceforge for every competition around the world.

On Sunday, after the finalists will have uploaded their code, we will
have another discussion session here in Atlanta and teams are invited to
send us more of their comments/suggestions.

Regards,

Gopal

On Sat, 2007-07-07 at 00:54 +0330, Hassan Nikaein wrote:
> Dear Cameron
> Thank you for your recent efforts at the games, we appreciate your
> work for the RCR community as you mentioned, we could give you some
> hints on the maps, we suggest that you see the link below which are
> the maps of Iran Open games:
>
> http://ce.sharif.edu/~nikaein/IranOpenMaps.tar.gz
>
> we wish you good luck at the games
> Impossibles team
>
>
> On 7/6/07, Cameron Skinner <cam_at_cs.auckland.ac.nz> wrote:
> Hi,
>
> > - My second comment is about the maps used these days. I
> think these
> > maps do not have required variety to evaluate team's
> behavior. You can
> > categorized most maps in 2 groups which has too many
> similarities. for
> > example:
> > 1) maps with many fires (not controllable fires - huge
> portion of
> > city burns for all teams) and too many blockades (whole map
> seems blocked).
> > 2) maps with some small fires (which most teams
> extinguish it easily)
> > and there are no considerable blocks.
>
> It is very difficult to generate new maps for the competition
> with the
> tools available at the moment. Unfortunately, the random map
> generator
> is very simple and we only have 4 other maps to use. In
> addition,
> generating situations that will be challenging is extremely
> difficult -
> it is hard to tell whether a particular fire will be
> controllable or
> not, or whether there are too many or too few blockades.
>
> I made some changes to the scenario generator yesterday that
> will allow
> for more interesting maps. Today we will hopefully see some
> more
> difficult situations that will test all aspects of each teams'
> performance.
>
> You are welcome to generate your own maps/scenarios and send
> them to me.
> In fact, I recommend you try it so you can get an idea of just
> how hard
> this problem is. If possible, I will try to include your ideas
> in the
> competition scenarios (obviously I won't just use your map
> since you
> have a team of your own, but any ideas will be appreciated).
>
> Regards,
> Cameron.
> _______________________________________________
> robocup-rescue-s mailing list
> robocup-rescue-s_at_cc.gatech.edu
> https://lists.cc.gatech.edu/mailman/listinfo/robocup-rescue-s
>
>
>
> --
> Hassan Nikaein ( Nikaein_at_Gmail.com )
> _______________________________________________
> robocup-rescue-s mailing list
> robocup-rescue-s_at_cc.gatech.edu
> https://lists.cc.gatech.edu/mailman/listinfo/robocup-rescue-s

-- 
Dr. Sarvapali Ramchurn
Intelligence, Agents, Multimedia Group    t: +44 (0) 23 8059 3270
Electronics and Computer Science          f: +44 (0) 23 8059 2865
University of Southampton                 e: sdr_at_ecs.soton.ac.uk
Southampton SO17 1BJ, UK.
http://www.ecs.soton.ac.uk/~sdr
_______________________________________________
robocup-rescue-s mailing list
robocup-rescue-s_at_cc.gatech.edu
https://lists.cc.gatech.edu/mailman/listinfo/robocup-rescue-s