[robocup-small] Ideas for 2007 Rules

[Krist Wongsuphasawat]

Dear all,

These are my opinions about proposed ideas.

- Move to a partially automatic ethernet-based referee box
>
>   I liked the demonstration by Plasma-Z on semi-automated refereeing.
>   To make such a system practical, we would need to move to an
>   ethernet-based referee box.  This has some other advantages: (1) no
>   more need for split-serial cables, and finding long ethernet cables
>   is much easier. (2) it makes it easier to verify teams are connected
>   to the referee using a "heartbeat" message. (3) it is much easier to
>   find computers with two ethernet ports now compared to an extra
>   serial port. (4) if the long term plan is to move to a shared vision
>   system, we will need to develop a shared ethernet-based
>   communications system anyway.
>
>   The challenge, of course, is transitioning to a new system, and
>   helping teams to correctly set up separate ethernet networks on
>   their computers (we do not want one shared system, as it opens up
>   all sorts of problems with communication interference).
>
>   We could try to develop an ethernet+serial referee box during the
>   transition period.  I would be willing to help out in that effort.
>

I agree with this idea and I like those many advantages that Jim said,
especially that we will need no more split-serial cables. But we'll have to
find out how to make sure that all teams will follow the rules and don't
cheat. As many teams asked me at the demonstration in Bremen last year what
will happen if any team always answers that another team made the last touch
before the ball run out of the field.

- Decrease timeout time to 5 minutes
>
>   Motivation: keep the games moving, and get them back under an hour.
>   Teams need to be ready for games.
>

I do not totally object this idea. But in my opinion, I think that 10
minutes timeout is quite reasonable already. Sometimes teams face situations
that something has to be done which need some time and I think that 10
minutes is not too long.

- Are chip kicks now too powerful for the goalie?
>
>   Proposal: allow a second defender to enter the defense area for a
>   limited time (3 seconds perhaps?)
>   Alternate Proposal: split the defense area into a small "no two
>   defenders" area, and a larger "no goalie touching" area.  This will
>   allow defenses a better chance of blocking while still preventing
>   walls across the entire goal.
>
>   Motivation: In the China open last week, for the first time two
>   teams capable of "header" shots met each other (CMDragons,
>   ZJUNlict).  After some early games adjusting things, we ended up
>   doing a TV demo that resulted in a 4-3 score after less than 8
>   minutes of play.  That projects out to a 16-12 or 16-15 score at the
>   end of the game, which is probably a little too high.
>
>   On the other hand, maybe corner kicks *should* be dangerous, just
>   like in real soccer.  If a team can complete too high a percentage
>   of corner kicks however, this can become a problem.
>

Well, I think if you really want to make corner kick more defendable,
the offside
rules may be more suitable as many people said. However, there are
difficulties in using the offside rules. We may not be able to decide
whether the robot is offside or not because they move too fast and as you
can see that many referee are new to the game and sometimes having problem
deciding the game with current rules. I can't imagine what will happen if we
ask those people to look for offside robots also. Automatic or
Semi-automatic system may have to be implemented to help solving these
problems.

- General kick speed limitation?
>   Maybe we should limit kicks to some reasonable upper limit, such as
>   10 m/s (CMDragons was using 15 m/s for the last three games).  What
>   is needed however is some way of enforcing this.  I don't have any
>   good ideas on this, but maybe someone else does.
>

I absolutely agree with this idea. There are so many reasons why we should
limit the power of the kickers to make sure that we are playing robot
soccernot war
machine soccer. First, I'm afraid that many teams will concentrate on making
their robot shoot faster and hope that their powerful shots would make the
ball penetrate through other teams' defence more than keep working on
strategies. Second, I can't imagine what will happen if the ball hit someone
near the field. That would be.... umm. I don't want to be that guy. We
should define the definition of safe kick and limit the kick speed to that
value.

- Increase team size to 6 robots
>   I think this might be the year to increase the number of robots, to
>   add more possibilities for passing, and to get closer to 11-vs-11 in
>   a manageable way.  Teams are scoring now, so we don't have the
>   situation of a few years ago with many 0-0 or 1-0 games.  However,
>   in order to prevent teams from stacking to defense too much, we
>   could *require* that at least one robot stay as a forward at all
>   times (i.e. a team must keep one robot on the offense side of the
>   field at all times).  This is kind of a reverse-offsides rule.
>
> - small field size increase (outer size unchanged)
>
>   With kicks now given 100mm from the border, the outer 300mm border
>   is not really necessary.  Thus, without any changes to vision
>   systems, we could decrease the outer border to 200mm on each side,
>   enlarging the field by 200mm in both dimensions.  In our lab, we
>   would not have enough space for this change, and this is probably
>   the case for many teams.  However, the size difference is not so
>   large, so teams should be able to verify easily that they can play
>   on such a field.
>

Bigger field is good but I think if we continue using the same number of
robots, we may have more space to play and solve the problem that sometimes
we feel that there are too many robots on the field than increasing the
number of robots also.

That's all for today.

Krist Wongsuphasawat
Plasma-Z, Thailand

-------------------------------------------------
Krist Wongsuphasawat
[ e - l i X i R ] StudiO™
part of YellowPiGz Production
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.cc.gatech.edu/pipermail/robocup-small/attachments/20061016/d94978b5/attachment-0001.html