Alt-BEAM Archive

Message #02749



To: beam beam@corp.sgi.com
From: Richard Piotter richfile@rconnect.com
Date: Wed, 28 Apr 1999 23:00:40 -0500
Subject: [alt-beam] Re: Contest!!!!


I think there ought to be two categories, battery, or solar (other "odd"
power sources could fall under solar).

Why???

I'm great with walkers, but I've yet to build a decent solar robot!
Other people have web pages stuffed top to bottom with solar engines and
photovores, and they don't have a working microcore! To lose judging
points for not being solar would realy limit the entrys. Two categories
would be nice, but not realy necesary. The judjing ahould be based on
how well it utilizes it's power, rather than what source it uses.

Well, that's my thoughts anyway! (:


Bob Shannon wrote:
>
> Steven Bolt wrote:
> >
> > On Tue, 27 Apr 1999, Chiu-Yuan Fang wrote:
> >
> > > Trust me! I tried. I couldn't think of a contest with your
> > > criteria and still make it accessable to newbies and vets alike.
> > > If you can come up with a good one (that can be honestly
> > > conducted on-line), I'm all ears.
>
> In Chiu's last contest we saw a number of new and innovative ideas
> dveloped for the competition in simply building a small, cool photovore.
>
> I think this next contest is a bit more ambitious, and will produce
> even more BEAM evolution than the last one did. Chiu's new contest
> seems to give the designer a lot more freedom to innovate!
>
> > Last time I checked, there were 344 subscribers on this list - not
> > significantly more or less than a year ago. The topics of
> > discussion seem also pretty much the same. One could even argue
> > that they have become less ambitious.
>
> Maybe because the lists ideas are firmly entrenched? There does seem to
> be some resistance to applying the BEAM philosophy to technologies that
> have traditionally been ruled out of bounds for BEAM design.
>
> > Obviously, there ought to be a contest for those who have just
> > discovered this hobby. But I think that a next, task-oriented level
> > would make BEAM more attractive for all, not just for those who'd
> > actually participate in that second contest.
> >
> > It should happen in an `arena', designed to restrict the size of the
> > `bots and provide them with an environment where cheap and simple
> > sensors are sufficient. The arena should not be too large - let's say
> > 100x50cm - and easy to put together. For a first `second level'
> > contest, a flat, bordered surface with simple obstacles might do fine.
>
> I orignally proposed using 'arena' comptition to compare and evolve
> behaviors and the body design of a standard development platform. At
> that time the general opinion on this list was that this was not the
> proper forum for that
> work. Despite other opinions, including the list owners, I choose to
> respect the general consensus and drop the subject here.
>
> Simply put, there is no way around needing to exchange and discuss
> source code.
> And I agree that this would not be appropriate here.
>
> BEAM traditionally uses the term' Jurassic Park'. I adopted the term
> 'arena' for a deliberately different plan for small robot evolution
> through competition. I would prefer that the term 'arena' not be used
> for a form of BEAM competiton, as this will only confuse my current
> efforts.
>
> Several people have already begun working on an Arena based robot
> evolution program, and it seems rather unfair for the BEAM list to first
> reject the discussion of this 'new approach' but then adopt the
> terminology.
>
> As a matter of mutual respect, please find another name.
>
> > And the task? For instance this:
> >
> > Five Alu foil balls with a diameter between 1 and 1.5cm would be
> > positioned at random. Within a certain time a robot would have
> > to find all these `Easter Eggs', and gather them within a 20x20cm
> > area. This `nest' might be marked with an omnidirectional IR beacon
> > and a black floor, the arena being white. Everything exept the
> > balls would be in a matt finish, making remote sensing of the
> > reflective (and conductive) balls at least possible. The organizers
> > would design and build a `bot which demonstrates that the task is
> > doable, and publish the details. The aim of the contest is to not
> > only do it, but do it faster. Points would also be earned by
> > performing the task with a simpler `bot.
>
> This sounds a lot like the MIT ants project.
>
> Is 'simpler' easy to define here? Simply using the component count
> would
> be unfair I think. Given the same parts count, would a BEAM design be
> judged as being simpler, or more complex than a CPU based design?
>
> > This particular task might stimulate developments useful for all
> > sorts of practical applications. The `bots need to rapidly explore
> > the whole arena, and have to make clever use of the single IR
> > beacon. They also must be energy-efficient to have sufficient speed
> > and stamina.
>
> It would seem that this would place a heavy penalty on solar robots.
>
> Rather than being efficient, you could simply buy an unfair advantage in
> the form of a Lithium Ion battery cell, and burn the milliampers up like
> they were free. A more powerful, lighter, but highly inefficient design
> can easily beat a highly efficient, less powerful design.
>
> You just made this a contest of battery technology. Let me enter one of
> the lithium sulfer based experimental cells, and you'll see what I mean.
>
> > Participants would test their robots in their own arena's. The
> > contest itself would use the `proxy' method, tried, tested and
> > still popular in indoor aeromodelling. You build your robot, write
> > up a short manual and mail everything to the organizers. A
> > competent proxy unpacks your robot, notifies you of arrival in good
> > condition, does minor assembly if necessary, perhaps charges its
> > batteries and finally puts it in the standard arena, but with
> > obstacles and balls in different, randomly chosen positions.
>
> Not a bad idea. Personally I'd hate to have to mail off my newest
> creations like that. (Are you sure you dont want to just swap code
> instead? Its much faster, cheaper and easier!)
>
> But if more exotic batteries are used, sending them off for charging
> becomes risky.
>
> > Afterwards the robots are mailed back together with any prizes won
> > and perhaps a video of the event. The results would of course be
> > published on the Web.
>
> A video (or CD) of the competition and judging would be great! This
> idea I really love. Heck, even a CD with all the web pages for all
> the contestants in the last competition sounds like a good idea as well,
> now that I think of it.
>
> With potential competitors being so widely distributed, I don't think
> that physically exchanging prototype, handmade robots is a viable idea.
> Its very expensive and the risks to the prototypes are not so small as
> to be ignored.
>
> Dammaged prototype's might not be recognised until after the events, and
> what protocol would permit a very expensive return, repair and re-trial
> in time for the final determination to be made?
>
> Heck, if I tried to ship off Vore-n-more, my wife would have a fit! I'd
> sooner drop out of the competition than risk the bitty bug on UPS's
> track
> record.
>
> Shipping alone could easily cost more than the prizes!

--


Richard Piotter
richfile@rconnect.com

The Richfiles Robotics & TI web page:
http://richfiles.calc.org

For the BEAM Robotics list:
BEAM Robotics Tek FAQ
http://people.ne.mediaone.net/bushbo/beam/FAQ.html

------------------------------------------------------------------------
eGroup home: http://www.eGroups.com/group/alt-beam
http://www.eGroups.com - Simplifying group communications

Home