[robocup-small] Rule changes for 2005, Commends from Cornell University

[Raffaello D'Andrea]

Dear Small Size League,

Please find our comments below.  These include feedback on Raul's proposal.
Raul, thanks for the in-depth and well written document.  

>Change 1:
>
>The dribbling distance will be set at 500mm for both active and passive
>dribblers.

We feel that we should simply eliminate the distance requirement.  Dribblers
only became a potential problem when Cornell developed side dribblers.  Horizontal
dribblers are a great way to obtain some control of the ball, but they are limited
in effect.  In particular, they are really only useful for getting control of a ball and receiving 
a pass. In 2003, when I was part of the rules committee, there were actually
three groups that felt we should simply ban the side dribblers, and only allow horizontal
dribblers: Lucky Star, RoboRoos, and BigRed.   

For those that want to see what the game is like with only horizontal dribblers, look
at the video footage from 2000 and 2001 (we actually had vertical dribblers in 2001, but
they were worse than useless.  They need to be at an angle to be effective).   The footage
speaks for itself: we haven't come close to that level of passing since that time.  If the field
had been larger in 2001, the passing game would have been outstanding.  

Our recommendation:  allow horizontal dribblers, ban all non-horizontal, moving dribbling surfaces. 
Simple, and to the point.  


>Change 2:
>
>For all restarts where the ball has gone out of bounds and the rules
>state the ball is too be placed ON a boundary line or within 10cm of a
>boundary line, the ball will be instead moved towards the middle of the
>field to an imaginary line that is 10cm any of the boundary lines that
>the ball is within the 10cm wall.  Needs rewording but I think it makes
>sense.

Reintroduce the 2003 walls.  They worked extremely well.  Getting rid of the sloped
walls was a huge mistake.  More on this later.



>Change 4:
>
>No more bluetooth is allowed in the small-size league, and all other RF
>decisions will be left to the OC based on local conditions.

Seems like an arbitrary decision.  What is the motivation?


>Change 6:
>
>The EC/OC/TC will decide upon ALL yellows, blues, greens, cyans, and
>pinks allowed to be used by robots before the competition and will make
>enough of these colors available at the competition for each team to
>field 6 robots.

What for?  Is this really a problem?  


>Change 7:
>
>We will have 6 robots per team

Again, where did this change come from?  See comments at the end.


>Change 9:
>
>There will be no chip kicks allowed during any restart.  Chip kicks (of
>any height) will be allowed during all other situations.

With regards to chip kicks.  I'm really addressing Raul's comments:  
We should simply enforce the 20 percent convexity rule.  I was not at the 
competition in 2004, but I was told that there were some teams that were blatantly violating this rule.
Scooping the ball up violates the 20 percent rule.  Chip kicking
can be achieved without violating the 20 percent rule.  For
example, the RoboDragons in 2003 had an excellent chip kicker.  Also note, however,
that it is not trivial to design a robust chip kicker.  It takes away from other design
choices, such as faster,  more maneuverable robots.  There is a tradeoff involved.

Addressing Sean's proposal. The rule could be the following:  treat all restarts, with the exception
of penalty shots, as an indirect free kick, with the caveat that the indirection
must come from your own team. In other words, the ball must touch
at least one other robot on your own team before going into the net.  

Overall comments:
----------------------
1) There were too many changes in one year.  The only major changes in 2004 should
have been to make the field larger and to ban side dribblers.    This is what we should
be doing for 2005.  We tried radical changes in 2004.  They did not work.  I think that
we will be in a much better position to address Raul's comments once we can see
what incremental changes lead to.   But I will comment on one of Raul's proposals:
battery/energy constraints:

2) Our game is much closer to hockey than it is to soccer. Or indoor soccer.   And in hockey, 
there are no such constraints:  the players go on, 
skate really hard for 2 minutes, and then take a rest for 3 to 5 minutes.  They go all out when 
they are on the ice. While I agree that putting energy constraints would make the problem interesting, it sounds too
complicated and unnatural, and is something that should be tackled in the simulation league, not a
hardware league.  Adding energy constraints will push the design space to lightweight, expensive
materials.  The robots will become much more expensive. Imagine what imposing the no-substitution
rule would have on hockey... 

3) In principle, I think that the automated referee is a great idea.  In 2001, we developed
our own competition, with real robots, to explore control of autonomous vehicles in uncertain
and dynamic environments:  RoboFlag.  For those interested, have a look at
http://roboflag.mae.cornell.edu/  In RoboFlag, there is no human referee.  It was
actually quite difficult to develop a set of rules that allowed human-free arbitration.
In RoboCup, we should work our way slowly towards this goal, with demonstrations
from several teams that want to take the lead. 



>_______________________________________________
>robocup-small mailing list
>robocup-small at cc.gatech.edu
>https://mailman.cc.gatech.edu/mailman/listinfo/robocup-small 


*****************************************
  Prof. Raffaello D'Andrea
  101 Rhodes Hall
  Sibley School of Mech. & Aero. Engr.
  Cornell University
  Ithaca, NY 14853-7501
  http://www.mae.cornell.edu/Raff
  (607) 255-0710 (Voice)
  (607) 255-1222 (FAX)
*****************************************