Alt-BEAM Archive
Message #11393
To: beam@sgiblab.sgi.com
From: Robert Stein rastein@dakotacom.net
Date: Mon, 06 Mar 2000 13:13:23 -0700
Subject: [alt-beam] Re: lobster
At 11:28 AM 3/6/00 -0600, you wrote:
> > Emergent behaviour is not precisely robot learning.
> > Connect
> > some
> > >mechanical parts to an electronic circuit, and produce
> > some
> > repeatable,
> > >robust behaviours under a variety of conditions -- that's
> > emergent
> > >behaviour (in my view -- subject to much debate, no
> > doubt).
> > This means we have to figure out some way
> > for
> > a robot to
> > >retain "learned" configurations when it is shut off.
>In a way, a robots life span is determined by its battery life. Each time it
>is turned on, it is like birth. When the battery runs out, that in effect is
>the death of the robot. In essence, if we recharge the battery and switch it
>back on, we have just reincarnated a robot. To allow it to retain information
>from its previous "lives" would be like us knowing we were the king of france
>in 1608 and we retain the knowledge of how to rule a country. A better way to
>approach this would be to find a way to prolong the lives of our robots.
>Remeber, some insects only live a day before they die. It seems our robots
>are the same way.
But do those short lived insects learn while they are here? That is the
question ...... learning or not. It (the bot) may still only be doing
exactly what it is wired to do, nothing more. I don't discount that it
interacts with its surroundings, but I am not so quick to say it is
learning. My car reacts to its environment, runs crappy when cold, runs
better when warm, but it is not learning, merely reacting.
just some food for thought
Robert
11394 Mon, 06 Mar 2000 16:26:35 -0400 [alt-beam] Re: lobster alt-beam@egroups.com michael.hirtle@ns.sympatico.ca (Michael Hirtle) Shouldn't we also consider Humans don't ever turn off until we die or go into a
comma,
If we go into a comma we have to relearn how to walk and more and all that
stuff.
Maybe to keep a bot from having to relearn all that stuff we would have to keep
the BEAM controller alive (the bicore, microcore.....) and just turn off the
motors.
That's what humans do when we sleep, we turn off our muscles when we go to
sleep.
instead of killing our bots (turning them off) we should make them go to sleep.
Bruce Robinson wrote:
> incorporate that behaviour into a robot, start it up from a random
> position, and watch it settle in to a controlled, systematic movement
> pattern -- that's motor learning, however primitive. Since our robots
> aren't able to retain the stable configuration the way we humans are,
> they have to re-learn it each time they start up.
>
> It's interesting to consider this in terms of more complex robots. Each
> time Walkman fired up, it would have to "learn" how to walk all over
> again. No problem ... it might flail around very briefly, but no harm
> done. Now try this with a bigger, much more complex robot, with several
> motors per leg. If it had to "learn" to walk each time, it may do some
> serious damage during the learning process, both to itself and its
> surroundings. This means we have to figure out some way for a robot to
> retain "learned" configurations when it is shut off.
>
> Bruce
Home