Re: [robocup-rescue-s] Complaint

From: <rahimi@ce.aut.ac.ir>
Date: Tue 09 May 2006 - 07:07:10 GMT

Dear Alireza,

I have a question about your arguments regarding the
modification on HP/Damage sensing (BTW, not as a member
of TC but as a [former] agent developer so do not
disregard my e-mail as an official response from TC).

> We think with current changes in damage and hp ambualanceTeams need to be
> wholy change. we think the solution should be completely change. and we
> think this is too late.

In what manner do you think that the solutions formerly
applied to ambulance teams should change? I think the
only [not so] major change would be on the 'death-time
estimator' module of ambulance teams. Suppose that you
had a simple reliable function like:

int estimateDeathTime(int hp, int damage, int currentTime);

which estimates the death-time of a civilian having the
information of its HP and damage in the time 'currentTime'.

if what you argue is about the reliability of its estimates
in the new situation, then calling the function like the
following form:

estimateDeathTime(hp - 50, damage + 4, currentTime);

returns a reliable death time in the sense that it returns
a value which is gauranteed to be less or equal to the
real deathTime.

So, simply modify one line of code in your agents and
you will still have a reliable and stable code. And no
such phenomenon as a civilian dying which is not expected
to die happens.

In general, the new problem would be on founding tighter
and more realistic bounds of 'a' and 'b' for the following
formula:

estimateDeathTime(hp - a, damage + b, currentTime);

Is it hard? Is it considered a major change? Just spend
a week studying and working on this problem and see the
results. In fact, this is not a problem that could be
solved by reverse engineering the simulator.

Note that the HP/Damage rounding feature has been added
since the release of package 0.49a. Was it considered
too late? If so, why do you let TC know your opinion
just now?

Regards,
Arash Rahimi

_______________________________________________
robocup-rescue-s mailing list
robocup-rescue-s@mailman.cc.gatech.edu
https://mailman.cc.gatech.edu/mailman/listinfo/robocup-rescue-s
Received on Tue May 09 10:39:02 2006

This archive was generated by hypermail 2.1.8 : Tue 09 May 2006 - 08:39:04 GMT