[robocup-legged] 11x11 RoboCup match

[Hans-Dieter Burkhard]

Hi Pedro,

you are totally right: scientific progress is the important point.
Therefore, it is not the problem to play just another game on another
field.

The problems should be:
How could robots from different teams play toghether?
How can we adapt to new conditions without knowing before all details?

Of course we have to check what AIBOs can do at all and to prepare for
their features.

And another important point (holds for all robot leagues): Real
cooperation will be necessary if we have more players and a larger field.
I hope to get new insights about that. The midsize test game in a Gym at
German AI conference las autumn was very promising (while still being
everything else but perfect).

Best regards

Diter


On Fri, 6 Jan 2006, Pedro M U A Lima wrote:

> Hi, all,
>
> My team is a newcomer to the 4LL, I am pretty conscious of all the
> technical difficulties involved, and also fully convinced that the
> amount of things we still do not know is perhaps even larger. Therefore,
> I might simply shut up and read your comments while our team learns more
> about the league challenges.
>
> However, even though I share some of the concerns raised here,
> especially those regarding limited time to change our software so that
> we cope with a larger fields, communication protocols to mix together
> robots from different teams, etc, of one thing I am sure: we should not
> be so worried about the impression we give to the media and so on, but
> we should rather be worried about whether we are fosterinc scientific
> innovations while we make our AIBOs play. And if we keep not facing
> problems like low light, larger fields, comm protocols, etc, we may play
>   nicer and nicer games, but we will become more of an soccer robots
> industry than researchers...
>
> Regards
> Pedro Lima
> (ISocRob-4LL, ISR/IST, Technical University of Lisbon)
>
> Oskar von Stryk wrote:
> > Dear 4L Teams,
> >
> > 11x11 is, of course, a nice idea which may draw a lot of public attention.
> > But if a lot of media is circulated around the mid size field watching
> > 11x11 Aibos then we would like to see some reasonable soccer like
> > behavior and do not want to have the media people shaking their heads
> > about the confusion on the field caused by Aibos without orientation on
> > the field. Videos documenting a bad 11x11 demo game may be connected
> > with the 4L league in the public awareness.
> >
> > On the other side no 4L team really can afford to spend a lot of extra time
> > on this special event, because credits are given only in the regular
> > soccer competition and the challenges.
> >
> > Therefore, the setup for the 11x11 demonstration must be defined
> > in a way to facilitate the adaption of existing software:
> >
> > - Lightning:
> >   Sufficient lightning for the Aibos on the whole middle size field
> >   and on any beacons around the field is a must.
> >
> > - Localization:
> >   Sufficent orientiation is also a must for a proper soccer game.
> >   The ML field dimension in 2005, see
> > 	http://www.idt.mdh.se/rc/Mid-Size/ (--> "Rules")
> > 	http://www.idt.mdh.se/rc/Mid-Size/Documents/msl-rules-2005-20050625.pdf
> >   was 8m x 12m, whereas the 4L field is 4m x 6m.
> >
> >   The width of the ML field is 2m longer than the total length of the 4L field !
> >   Therefore it will be extremely difficult to use
> >   ML beacons for localization.
> >
> >   The ML field line width is <= 12.5cm in the rule book. Some lines may be
> >   detected but corners or edges may be difficult to see from many positions
> >   on the field as well.
> >   Localization based on the lines only may also be very difficult.
> >
> >   Therefore it is strongly advisable to use additional aids,
> >   like several additional beacons around the ML field for the 11x11 demo.
> >   Why not using at least 4 (or 6) additional beacons on each side
> >   and also some on each goal side?
> >
> >   --> We need a proper definition of the 11x11 demo field layout
> >  	with the field dimensions, the lines and line width
> > 	and the position and colours of additional beacons for orientation.
> > 	Who will be taking care of this?
> >
> > - Communication:
> >   If not just the software of one regular 4L team is used on all 11 Aibos
> >   of the 11x11 demo game, then we need to define, e.g., the data
> >   to be exchanged between Aibos of different teams enabling team play.
> >
> > --> Who is taking care of organizating the required definitions
> >     (and proper installation in Bremen)?
> >
> > It would be very helpful if the 4L Technical Committee could take care of
> > this and possibly publish a document describing the modified  ML field setup
> > for the 11x11 demo after discussion on the mailing list.
> > It would also be helpful if all teams could know this setup early, e.g.,
> > by the end of February.
> >
> > Btw, does the current 4L game controller already support
> > two teams of 11 players?
> > And labels with numbers from 1 to 11 (14 incl. substitute players)
> > are needed for each team.
> >
> > Best wishes to a happy and successful new year to all 4L teams!
> > Looking forward to seeing you in Bremen!
> >
> > Oskar von Stryk
> > Darmstadt Dribbling Dackels, Member of German Team
> > www.dribbling-dackels.de
> >
> > On Tue, 3 Jan 2006, Thomas Röfer wrote:
> >
> >
> >>>As a bystander in this discussion, I am just wondering about
> >>>the proper illumination. To the best of my knowledge, the
> >>>mid-size field is illuminated
> >>>insufficiently for AIBOs to operate with their poor visual sensors.
> >>>
> >>>Is it really an issue? If yes, will it be addresed?
> >>
> >>Hi,
> >>
> >>since it is also planned to have four-legged games on the center court (on a
> >>normal four-legged league field), there has to be some extra lighting. I
> >>believe if we ask for sufficient lighting for the entier MSL field, we will
> >>get it (for the center court only).
> >>
> >>Best regards
> >>
> >>Thomas Röfer
> >>
> >>
> >>------------------------------------------------------------------------
> >>
> >>_______________________________________________
> >>robocup-legged mailing list
> >>robocup-legged at cc.gatech.edu
> >>https://mailman.cc.gatech.edu/mailman/listinfo/robocup-legged
>
>
> --
> Pedro U. Lima
> Instituto de Sistemas e Robotica,
> Instituto Superior Tecnico - Torre Norte
> Av. Rovisco Pais, 1
> 1049-001 Lisboa
> PORTUGAL
> Tel. +351-21-8418274
> Fax  +351-21-8418291
> http://www.isr.ist.utl.pt/~pal
> _______________________________________________
> robocup-legged mailing list
> robocup-legged at cc.gatech.edu
> https://mailman.cc.gatech.edu/mailman/listinfo/robocup-legged
>