AW: [robocup-small] Rule changes in SSL (automatic ref and batteryswapping)

[Raul Rojas]

My answers:

>I find this "energy budget" of an automatic ref an interesting concept.
I 
>have some questions about the idea.

>Would the energy budget be global to the team?  

No, I propose individual energy budgets, like in real soccer. By the
way,
the idea of constant energy per robot came from Team Canuck, I think in
2002,
when they proposed to consider a standard battery. However, I think that
a hardware
solution is more difficult to implement than a computer vision solution.
Just measure the kinetic energy spent by a robot, and it will correlate
pretty good with the energy drained from the batteries.

>how does one account for arbitrary movements of a robot when its 
team mark may be occluded by a handler etc?

That does not count, only when the robot is being tracked by the vision
we record the energy spent.

>Does the energy budget build over time?  

No, each robot has a maximum amount of energy E. If the robot runs
forward with speed 3, it consumes nine times the energy than if he runs
with speed 1. So, it pays to run fast when something important is
about to happen, but it pays to come back slowly, to save energy.
This is just an example. Any kind of energy function can be used
(quadratic,
cubic, in-between).

>How would a team be penalized for  going over the energy budget?  

The robot cannot move, it has exhausted its energy. 
What do you do when you are about to lose a robot? You let it play
defense,
probably.

>How would an automatic ref account for substitutions?

The new robot gets the remaining energy of the robot which went out.


But please remember, this is all about 2006 or beyond. 2005 we won't
have
anything of this, not with vision, not with standard batteries or
similar
ideas in 2005. That is too much of a change for 2005 (I think).

Raul Rojas
FU Fighters Team




Mark Goldman
Ohio University, RoboCats

The views and opinions expressed are my own and not, necissarily, those
of 
my team.

>Just a comment about battery capacities. The real capacity can differ 
>sometimes by up to 20% from the nominal capacity. It also changes with 
>the age of the battery. So, what you want would be nice, but in real 
>life some teams would end up having 20% more energy than others, 
>depending on battery brand.
>
> >(c) Limit the battery capacity
> >(d) No replacing of batteries allowed during the game
>
>That is why my proposal to use an all-electronic referee in 2006 (I 
>repeat 2006) includes the idea of doing some form of energy management,
>purely using the vision software and not by hardware.
>
>Measuring energy consumption on the robot itself using hardware could 
>be also done, of course, but could lead to horrible discussions ("your 
>measuring hardware is giving incorrect readings", etc.).


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