[robocup-worldwide] The essence of soccer

Thu Nov 5 09:35:48 EST 2009

Dear Peter Stone, Dear Colleagues,

1. Thanks for this inspiring initiative, of setting concrete rules for  
mixed, robot-human, soccer.

2. I advocate reasoning in a way that starts from the goal !

3. Goal: We all wish pleasant competitions where success cannot be  
forecast for either sides.

4. As for most (all ?) popular games, in soccer too, there is in  
practice a certain amount of stochastic factors, that, per se, helps  
fullfilling the wish expressed in point 3.

5. Let’s nevertheless neglect those random factors in the sequel.

6. If by some magic a « perfect » set of rules was created, whereby  
machines could be opposed in a « fair » way to humans, this would not  
be very interesting : except for a short transition time, maybe, say,  
in 2039 (optimistic example) or maybe 2072 (pessimistic example) the  
result of human-robot soccer games would be predictable. i.e. without  
interest.

7.   I think it would be much more interesting to balance predictable  
advantages by a specific strategy. Classical approaches of this type  
include the following ones:
7.1 - giving regularly revised handicaps (e.g. n robots against m  
humans ; or goal width of different size ; or adding a n kg load; or  
adjusting the amount of allowed power ; or limiting the amount of  
cognitic performance levels; etc.)
7.2 – classifying teams by performance levels (e.g. our robots should  
start in more modest official leagues than in the world final cup!;   
and later on play among themselves ?)
7.3 – balance the team levels by dynamically exchanging players !  
(e.g. in popular soccer, if a player scores a goal, he/she then goes  
on playing for the opposite team – notice that this tends to balance  
the values of teams also when a player scores againt his/her own  
team !). By this token, we might have mixed robot-human teams.
7.4 – A most interesting case which overlaps with 7.1 and 7.3, would  
be to have a mixed human-robot team. In terms of handicap, in early  
years, adding a robot instead of a human player would be adding a  
positive, incremental handicap for the team; and hopefully soon the  
contrary should apply ! Notice that by this token, as a Federation,  
Robocup should start soon organizing its own reference human soccer  
player team !

8. Two nice side effects of applying some ideas of point 7 include the  
fact that we could start those soccer competitions much earlier than  
in 2050 ; and that this development would immediately connect with  
what we are doing in technical leagues right now, as shown in the  
short examples at the beginning of Peter's paper.

9. In conclusion, I would like to make explicit what probably eveyone  
thinks : the main, crucial rules to define are those tending to ensure  
the safety of soccer players, humans and robots, during competitions,  
on the playing field, as well as around it for the public, and before  
and after the game, in training and other ancillary phases.

10. And for me, in as much as robots can be part of the game, they can  
play soccer.

Best regards,

Jean-Daniel

Le 4 nov. 09 à 03:11, Peter Stone a écrit :

>
>  --------------------------------------------------
>  [[ Apologies if you receive this multiple times ]]
>  --------------------------------------------------
>
> RoboCup soccer colleagues,
>
> I'm writing to let you know about an article that I recently wrote  
> with
> Michael Quinlan and Todd Hester on "The essence of soccer" that is now
> available from my website:
>
>     http://www.cs.utexas.edu/~pstone/Papers/bib2html/b2hd-PHILOSOC10.html
>
> It will appear in the book "Soccer and Philosophy" sometime before the
> world cup.
>
> In it, we list a proposed set of rules for a match between humans and
> robots that should lead people to agree that robots are better than
> people if they win.
>
> I truly believe that because of all of your/our efforts, that that  
> match
> will happen by the year 2050.  If so, I hope that the article will  
> serve
> as a good starting point for the rules of the match.  At the end of  
> the
> article, we acknowledge all RoboCuppers from your contributions.
>
> Please feel free to share this article with anyone who might be
> interested.
>
> Best,
> 	Peter
>
> ___
>
> Peter Stone, Associate Professor
> Department of Computer Sciences    phone: 512-471-9796
> The University of Texas at Austin  fax:   512-471-8885
> 1 University Station C0500         pstone at cs.utexas.edu
> Austin, Texas 78712-0233  U.S.A.   http://www.cs.utexas.edu/~pstone
>
>
> _______________________________________________
> robocup-worldwide mailing list
> robocup-worldwide at cc.gatech.edu
> https://lists.cc.gatech.edu/mailman/listinfo/robocup-worldwide

---
Prof. J.-D. Dessimoz, ing. et Dr ès Sc. techn. EPFL, MBA UniL

HEIG-VD // Haute Ecole d'Ingénierie et de Gestion du Canton de Vaud //  
School of Business and Engineering
HES-SO // Haute Ecole Spécialisée de Suisse Occidentale // Western  
Switzerland University of Applied Sciences
Route de Cheseaux 1, CP, CH-1401 YVERDON-LES-BAINS, Switzerland
Tél.(direct):+41(0)24 55 76 315  Mobile:  +41(0)79 656 1691 (secr.): 
+41(0)24 55 76 405 / 318
Fax (secr.):+41(0)24 55 76 320 Fax (JDD/JDZ): +41(0)24 420 3234
e-mail : Jean-Daniel.Dessimoz at heig-vd.ch,  Jean-Daniel at Dessimoz.org,   
URL: www.iai.heig-vd.ch  http://lara.heig-vd.ch  www.heig-vd.ch
Autres relations: iAi - ISYS; LaRA-Robot CH – Eurobot, Robocup-at- 
Home; Manufuture; SwissSpace – IAF, PAC, SEOC ; CeTT, SwissT.net.

Recommendation. In case of difficulties in displaying this email try  
the following control: Message - text encoding - Unicode (UTF-8).
____________________________________________________________________________________________________________________________

DISCLAIMER :   This message is intended only for use by the person(s)  
to whom it is addressed. It may contain information that is privileged  
and confidential.
If you are not the intended recipient of this message, kindly notify  
the sender immediately and destroy this message. Thank You.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.cc.gatech.edu/pipermail/robocup-worldwide/attachments/20091105/07394ef3/attachment-0001.html 