Subject:
|
Re: Idea for a competition
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Thu, 8 Dec 2005 19:35:22 GMT
|
Original-From:
|
Chris Magno <{cmagno@rogers.}ihatespam{com}>
|
Viewed:
|
1225 times
|
| |
|
|
The idea for a "Block Transfer" style game has been proposed and will be
tried in Feb. 2006 for the next rtlToronto game R20 aka Project Y
http://peach.mie.utoronto.ca/events/lego/lego-rtl20/
The basic idea is to start with ax 8' x 8' play area with 6" walls, and
have up to 10 autonomous robots each with a "standardized" vertical
hole/hopper. The robots goal is to transfer a 2x2 "block" into the
hopper of another robot. As well as having a robot who can accept
blocks from other robots. The score is how many blocks you have
successfully transfered multiplied by how many blocks you have received.
Each robot will have a limited number of blocks. We are going to try
the idea of limiting the number of blocks a robot can have to the number
of robots in the play area. (so if we have 7 robots, each robot will
start with 7 blocks.) Each robot will have its own colour of block to
start.
I know its not exactly like the idea that Ignacio put forward, but we
are going to give it a shot. The real challenge here is both technical,
and social. We think that, unlike every other rtlToronto robot game
to date, it is in the best interest of the humans to publicly share as
much info about the robot as they can. We are referring to this aspect
of the game as the "social engineering" portion. I am curious how well,
ideas and "standards" will "evolve" based on the idea that someone
might tell others that
"... my robot will have a light mounted at X and Y, and when it is on,
my robot will be stopped"
Then, people are free to choose to also copy this idea, or design for
looking for a light at X and Y.
I put together 2 very simple robots to try to test how a robot might
seek another.
-----------snip----------
this is a BAD web cam compressed 320x240 AVI of 16 seconds (1/2 meg) of
robot seeking.
www.thepyroguys.com/videos/robot-seek.avi
what you might see, is that both robots are started at the same time.
they both are programed to have a 50/50 chance of either turning on a
beacon and waiting for x seconds, OR not turning on a light and seeking.
the robot on the left in this video is seeking, and the one on the right
has chosen to stay put, and turn on a beacon light.
-------------end--------
Chris
Mr S wrote:
> Ignacio,
> That is an interesting concept. What happens if a
> robot runs out of 'things' to trade?
>
> Are there rules on when or why a robot would trade? or
> be forced to trade?
>
> What if they are sent on 'course' where to get from
> starting point to finish point, they would have to
> encounter other robots, thus be in a position to trade
> or not, and based on trading status, their progress is
> affected somehow. This would give a goal to the
> competition, and use the trading situation as a
> penalty kind of thing. Say the course is a long line
> following course, and to get another bot out of the
> way, you have to pay a token, or they pay one to not
> get out of the way?
>
> The robot that finishes the line course first with the
> most tokens wins?
>
> I don't know how that would work as a contest, but it
> would add more interaction to the standard line
> following courses.
>
> Just a thought...
>
> --- Ignacio Martinez Vazquez <ignamv@gmail.com> wrote:
>
>
> > Hi,
> > Here's an idea for a competition. Unfortunately, it
> > requires several robots,
> > meaning you either need to have several RCX/board of
> > choice, or know someone who
> > does robots.
> > There's an area, delimited by a line, in which the
> > bots (more than 2) wander
> > around until they recognize another one. Then, they
> > read a color strip (like
>
>
> <http://www.philohome.com/cardreader2/cardreader.htm>)
>
> > and decide whether to
> > separate, or continue. If they continue, they trade
> > something. Based on some
> > criteria, they may decide not to give the other one
> > the 'thing' (meaning they
> > trick the other into giving away a thing). Each bot
> > has several of them.
> > However, after finishing the trade, they modify each
> > other's cards. If
> > programmed to associate certain sequences with bots
> > who give or scam, they
> > become more likely to earn things.
> > Problems:
> > * Cards would be a total bitch to read and, most of
> > all, modify. They could be
> > transmitted and received through IR, making it
> > easier.
> > * Trading would also be hard. A simple mechanism
> > would be necessary, since
> > probably 2 motors are being used to drive. Maybe the
> > things could be technic 1x2
> > bricks, held on axles. However, aligning axles and
> > actually pushing the bricks
> > would be hard.
> > * Kinda boring, specially if someone didn't plan to
> > recognize scammers and runs
> > out of bricks at the beginning.
> > I guess IR could be used so that, when 2 bots are
> > pointing at each other, they
> > both receive each other's cards. That way, if they
> > both want to trade, they just
> > go straight on. Otherwise, they look away. You'd
> > have to make sure one bot isn't
> > getting messages from more than 1 other at the time.
> > I know it's not great, but I just thought it up...
>
>
>
>
>
|
|
Message has 2 Replies:
 | | Re: Idea for a competition
|
| "Chris Magno" <lego-robotics@crynwr.com> wrote in message news:43988AFA.40006@...ers.com... (...) This AVI is corrupt or improperly named. Windows Media Player here plays it as a sound file. Cheers ... Geoffrey Hyde (19 years ago, 9-Dec-05, to lugnet.robotics)
|
Message is in Reply To:
| | Re: Idea for a competition
|
| Ignacio, That is an interesting concept. What happens if a robot runs out of 'things' to trade? Are there rules on when or why a robot would trade? or be forced to trade? What if they are sent on 'course' where to get from starting point to finish (...) (19 years ago, 8-Dec-05, to lugnet.robotics)
|
8 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
