To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 11054
11053  |  11055
Subject: 
Re: A modest proposal/challenge/whatever: 'bot wrestling
Newsgroups: 
lugnet.robotics
Date: 
Sat, 18 Mar 2000 20:59:58 GMT
Original-From: 
Brian Connors <CONNORBD@YAHOOnomorespam.COM>
Viewed: 
4345 times
  
--- Doug Weathers <weathersd@metro.dst.or.us> wrote:
> in article
> 20000317185227.4930.qmail@web906.mail.yahoo.com,
> Brian Connors at
> lego-robotics@crynwr.com wrote on 3/17/00 10:52 AM:
> 
> > Lego Wrestling, anybody?
> > 
> > It's sort of a natural extension of something like
> > Robotag and is vaguely inspired by the WWF's • Brahma
> > Bull Rope match.
> 
> WWLM? :)
> 
> > Basically, two robots get in a
> > circular, one-meter-wide ring, are lashed together
> > with a plastic or twine lanyard about a quarter • meter
> > apart (so they don't have to worry about finding • each
> > other),
> 
> There should be a standard for how the lanyard
> should attach to the robot.
> A standard assembly of some kind.  Perhaps it should
> incorporate a swivel
> and be mounted on the top.  Or do you just want to
> have it hooked to the
> side of the robot?

Actually, that's part of the challenge: designing a
legal lanyard support that won't either intentionally
break off or get tangled. I sort of envision it being
attached to opposite ends of a touch-sensor bumper,
but that's not the only possibility.

> > and then try to either throw each other out of
> > the ring (a la sumo)
> 
> It would help if the robots knew the limits of the
> ring.  Perhaps a warning
> stripe painted on the floor a few inches inside the
> actual perimeter of the
> ring.

One ahead of you on that one. The way I envision it,
the ring actually has a 50-cm-diameter white center
and a 25-cm red outer zone, idea being if you're in
the red zone and moving backwards it's time to start
the motors.

> > or immobilize/incapacitate each
> > other.
> 
> Who wins in this circumstance?

Ref's call. An obvious submission maneuver equals a
win, as does any clear indication that one 'bot did
initiate the situation. If it's unclear, it's declared
a stalemate.

> > Winner is last 'bot standing after thirty
> > seconds.
> 
> I think that thirty seconds is a bit too short of a
> time limit.  Give the
> poor things time to figure out what to do!  I'd
> suggest that if nobody has
> won within two minutes then the match is a draw.

That's flexible, but with such a small ring it's
likely that if a ringout doesn't happen within the
first thirty seconds the match is stalled anyway. See
above for details...

> > 
> > Thoughts? I'm doing a writeup right now on one way • of
> > doing it, and I'll have it up next to the • (shameless
> > plug) What Set Should I Buy FAQ at
> > 
> > http://www.geocities.com/connorbd
> > 
> > early next week.
> > 
> > /Brian
> > 
> > =====
> > --
> 
> 
> In the MIT Lego competitions, individual robots have
> to qualify by being
> able to beat an inert block.  If they can't score
> higher than the block,
> they probably won't do too well against a real
> opponent.

:-)

Shouldn't be much of a problem here; the block either
gets whipped out of the ring with a sort of drag-halt
maneuver (a whip) or simply pushed out (a slam). Of
course, practicing a submission-type maneuver (a
"chain" or an abbreviated drag) won't be much good
against an inanimate target, but you don't expect a
punching bag to hit back either.

> How would you qualify a RoboRoper bot?  What sort of
> inert block would you
> use?

> How would you compete against a heavy bot with
> rubber treads that won't drag
> worth beans?

Ram it. Over and over again until something falls
apart if necessary. (That's the "slam" maneuver I
mentioned above)

> I don't see a good way to make a robot that can
> upset the other one by
> yanking on the rope.  Trying to tow the opponent out
> will result in the In
> my mind, this turns into a sumo competition.  As far
> as I can see, the rope
> can only be irrelevant (pushing the opponent out) or
> cause a tie (entangle
> both robots).  Maybe I missed something and need to
> spend more time thinking
> about it.

It depends on what you consider a win. It is actually
part sumo in conception (mainly because I can't see a
robot being smart enough to do much else), and a
simple drag actually results in a self-DQ because the
dragger leaves the ring first. Even if that's not
considered the equivalent of an own goal, it simply
reduces the whole thing to a sort of tug-of-war.
That's not that fun or creative.

> Now, if you change the rope to a rigid rod, things
> start to get more
> interesting.

*That's* very true. I suppose we could have two types
of matches: rod and lanyard. (Freestyle is still out,
unless you have thoughts of playing with light
sensors.)

/Brian

=====
--



Message has 1 Reply:
  Re: A modest proposal/challenge/whatever: 'bot wrestling
 
If you need further inspiration for this idea, get hold of a copy of Jack Vance's SF book "The Face" and read about Haudal, a game where some number of players start out in a central yellow circle, which is surrounded by a larger green cirle, which (...) (27 years ago, 19-Mar-00, to lugnet.robotics)

2 Messages in This Thread:

Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR