Re: rules for 2002 (version.7) & official registration & award

From: Mazda Ahmadi (
Date: Wed 01 May 2002 - 15:54:58 GMT

Dear Tak,

I agree with you that there should be a restriction on the length of the
messages. But I believe it's too late for setting such a huge
restriction. I don't think that there have been enough discussion on
12kkID, It's a big element in Rescue Simulation and there should be
enough investigation on this issue.

Comparing with RoboSoccer, They have set their new rule about three
months ago, So the teams had enough time to design their teams.
And the other main difference is that every cycle in RoboSoccer
is simulation of 1 second in real world. But in RobocupRescue each
cycle is simulation of 1 minute. I think sending only 4 messages in every
cycle with 12kkID format is a low communication even for real humans.

I hope we can set a percise and good rule for 2003 with help of related

Best regards,

>From: (Ikuo Takeuchi)
>To: r-resc@ISI.EDU
>Subject: Re: rules for 2002 (version.7) & official registration & award
>Date: Wed, 1 May 2002 18:18:19 +0900 (JST)
>Dear All,
>We have a serious concern about the vagueness of the 2002 rule as was
>decided recently (version 7). There seems to be no clear criteria in
>advance to what extent we can allow the amount of messages issued by
>an agent at one cycle. The definition of a 'simple' sentence is one of
>the most important issues when we think of the RoboCupRescue
>competition as a good research platform for distributed multi-agent
>In our proposal made in last November,
>a 'simple' sentence is ultimately small with an expectation that such
>simple sentence makes the problem more challenging in the field of
>multi-agent systems research. "Not very large but some amount of
>messages" does not give a very good, clear-cut means or standard to
>measure the obtained ability of agent cooperation.
>RoboCupRescue competition is not a mere competition, but it should
>promote the related research to some good direction. In RoboCup
>soccer, recently, the size of a message issued by a player is
>drastically decreased than before to be nearly equal to that of a
>human player. We think that was a right decision. In RoboCupRescue,
>however, the situation is not so straightforward since future Robot
>rescue agents need not be restricted to speak as slowly as human
>agents. Too restricted 'simple' sentence may not match our final
>But we have not yet real Robot resue agents at present. So we think
>it is natural for the rescue agents in the current competition to
>model human agents as they are in order to suggest the real human
>rescue agents how they should cooperate in a fuss of big disaster. If
>we do not think so, that is, if we think the rescue agents should
>model futuristic robot agents in 2050, we should not restrict the size
>of messages at all! "Not very large but some amount of messages" does
>not seem to have a firm principle with respect to this kind of
>Moreover, we proposed this ultimately simple sentence rule about five
>months ago on the community. It has been long public enough to
>Best Regards,
>Tak (NUE)

Send and receive Hotmail on your mobile device:

This archive was generated by hypermail 2.1.4 : Sun 22 Sep 2002 - 18:52:58 GMT