Alt-BEAM Archive
Message #05775
To: beam@sgiblab.sgi.com
From: JVernonM@aol.com
Date: Wed, 18 Aug 1999 08:15:12 EDT
Subject: [alt-beam] Re: Contest entry (the bounty bot)
OK, Rich. I've seen the diagram before. I wanted to see a real, working,
tested robot. I don't mention or post one until it passes all of those
criteria. As far as I'm concerned this bot does not yet exist. It is a
drawing. I have literally hundreds of bots at that stage. But, if you check
my site, you'll find only finished, working bots. As far as I know, that
drawing is a couple of months old now. And someone actually backed out of the
contest because of a drawing? Amazing. By the way, wasn't it you who wrote me
a book on not sending such large files to the list? I'll be tactful enough to
keep mine short.
Jim
5776 Wed, 18 Aug 1999 12:05:08 -0400 [alt-beam] Overcoming obstacles - an idea Beam "Sathe Dilip"
--------------ECECFB10A2C20D9C1517081B
content-transfer-encoding: 7bit
content-type: text/plain; charset=us-ascii
I was thinking on the following lines even before the contest
(Innovative motion) was announced. Wilf's comments in an article last
week prompted me to write about it. I hope someone finds it interesting
enough to build a bot based on it.
I will explain the general idea and leave the details to the builder. I
do not know if someone has tried this before.
The idea is to build a bot which combines some features of the
photovore, (the miniball to some extent) and the walkers. I haven't
built any yet but based on my understanding of how they work, photovores
seem to have a problem negotiating stepped surfaces. Walkers can get
around these to some extent.
My idea is to use two motors (one on each side) and spokes in place of
wheels. 4 wires/rods at 90 degrees each should do. Due to the torque
needed to move the spokes/legs, we will need geared motors. The
similarity to miniball comes from the way the motors/circuitry/power
source is suspended from the legs. Unlike the walkers the legs turn
full circles (like wheels). However due to the spacing between the end
points (on circumference) where the legs can come to rest on a surface
(unlike wheels), they should be able to negotiate the stepped surfaces
better.
At any given time, two legs on one side (and motor) of the bot will
remain inactive while the other side pivots over one leg and advances.
So the movement will be somewhat of a tumbling nature (may be we should
call this guy TUMBLER?).
A further development can be tried to make the robot more adaptive to
its environment. The bot starts with a given leg length (shortest -
just enough to clear its underbelly thus requiring less torque and power
to move about). If it finds that it can not cross a step, it extends
its legs (lehthens them - all 8 spokes at the same time for simplicity
and balance). Then it again tries to cross the obstacle. The
lengthening of legs could be a one step or multi-step process. It
should be possible to achieve the lengthening/shortening of legs through
the use of a third motor or a stepper mechanism (a relay with a dog
wheel/rachet?) for activation. Actual lengthening can be achieved by
either a rack & pinion mechanism or with two concentric rings and levers
(think of the old Compur shutter mechanisms in some cameras- only inside
out).
I have attached a very bad graphic representation of the tumbler & how
to implement the leg extension. I hope the 4K gif file is not
objectionable. The leg shape shown, allows one to choose between a
rolling and a tumbling gait. The claw shape might help in getting a
good foot-hold on steps (especially when augmented by ribbed rubber
boots).
While writing this, I started thinking. May be for a future
competition, we can collect new ideas. Then base the competition on
building better bots based (I like the sound of that :-) on one or more
chosen ideas. We can also have people collaborate so that people (like
me) who do not find the time to build can work with someone who does or
those with ideas needing special building skills/facilities can work
with those who have access to such facilities.
This is getting too long so I will stop. Thanks for reading. Let me
know what you think.
Dilip
--
Pl. remove *s from the e-mail address to reply
--------------ECECFB10A2C20D9C1517081B
Attachment: Tumbler.gif
--------------ECECFB10A2C20D9C1517081B--
5777 Wed, 18 Aug 1999 11:20:47 -0500 [alt-beam] Re: Contest entry (the bounty bot) beam@sgiblab.sgi.com Richard Piotter It appeared as a broken file when I saves it, draged it, and tried it
with a jpg viewer or Stuffit. It definitely didn't come through over my
Macintosh. Oh well. That's why I say away from MacroCo$t... Err...
Micro$oft products. On the TCRG group list (Twin Cities Robotics Club)
we got a guy provlaiming the woes of how he couldn't instal Visual Basic
6.0 (I think) properly cause it forced him to install explorer, and it
actualy made it so netscape would not run cleanly... Now I know what
this federal case against Microsoft was about!!! Holy crud!!! They then
got into a discusion on how they would never want to see a 300 pound
robot headed for them runing windows, cause they would fear for their
lives! Man! It was great hearing that :)
Check your mail settings maybe. Stupid microsoft! Outlook may have more
features, but I stick with simple, non bloated Netscape Messenger mail box.
Ben Hitchcock wrote:
>
> well,
>
> >>
> >> I made a solar powered robot similar to LEM about 5 months ago ...
> >> IF anyone's interested, I'll post some pictures.
> >
> >Of course we're interested.
> >
> >Bruce
>
> you asked for it...
>
> I dusted it off thisafternoon, fired up the Quicktake, and took these two
> pictures.
> The one on the left is a shot of it sitting on top of a millimetre ruler,
> and the shot on the right is the underneath of it. You can see the three
> motors mounted photopopper-style with fuse clips, as well as the 0.047 uF
> cap, and the underside of a 74HC240.
>
> Let me know if anyone can't view it properly - I think it gets binhexed, but
> Outlook Express isn't very clear on what it actually does to attachments.
> Macs should read it fine. (And who cares about anyone else? Sorry, couldn't
> resist ;-)
>
> Ben
>
> ------------------------------------------------------------------------
>
> tricycle.jpgName: tricycle.jpg
> Type: Macintosh BinHex Archive (application/mac-binhex40)
--
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
5778 Wed, 18 Aug 1999 11:25:39 -0500 [alt-beam] Re: Contest entry (the bounty bot) beam@sgiblab.sgi.com Richard Piotter How long was it? According to my computer, it was only 14.6 K!
That is not a giant file. I've seen 8K text messages before! Also, if
any person is inspired to build such a robot because of the image, then
it has already served a purpose! I LET people know what I'm doing, so
they can try to do their own independent project if that is what they
choose to do!
JVernonM@aol.com wrote:
>
> OK, Rich. I've seen the diagram before. I wanted to see a real, working,
> tested robot. I don't mention or post one until it passes all of those
> criteria. As far as I'm concerned this bot does not yet exist. It is a
> drawing. I have literally hundreds of bots at that stage. But, if you check
> my site, you'll find only finished, working bots. As far as I know, that
> drawing is a couple of months old now. And someone actually backed out of the
> contest because of a drawing? Amazing. By the way, wasn't it you who wrote me
> a book on not sending such large files to the list? I'll be tactful enough to
> keep mine short.
> Jim
--
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
5779 Wed, 18 Aug 1999 12:20:18 -0700 [alt-beam] Re: Overcoming obstacles - an idea alt-beam@eGroups.com jon_com@hotmail.com jon_com@hotmail.com wrote:
I had thought of somthing similar to this, although in far less detail.
I would like to build something (one day) using your idea. I am new to
BEAM robotics and need to build my first robot (waiting for parts)
first. So it will be a little while before I do anything (2 yrs
maybe???) Anyway, great idea. Keep up the good work.
Jonathan Smolders
P.S. I just joined this mailing list. If anybody want's to send me BEAM
related E-mail, please send it to instead of
my other e-mail listed (jon_com@hotmail.com)
original article:http://www.egroups.com/group/alt-beam/?start=5776
> I was thinking on the following lines even before the contest
> (Innovative motion) was announced. Wilf's comments in an article last
> week prompted me to write about it. I hope someone finds it
interesting
> enough to build a bot based on it.
>
> I will explain the general idea and leave the details to the builder.
I
> do not know if someone has tried this before.
>
> The idea is to build a bot which combines some features of the
> photovore, (the miniball to some extent) and the walkers. I haven't
> built any yet but based on my understanding of how they work,
photovores
> seem to have a problem negotiating stepped surfaces. Walkers can get
> around these to some extent.
>
> My idea is to use two motors (one on each side) and spokes in place of
> wheels. 4 wires/rods at 90 degrees each should do. Due to the torque
> needed to move the spokes/legs, we will need geared motors. The
> similarity to miniball comes from the way the motors/circuitry/power
> source is suspended from the legs. Unlike the walkers the legs turn
> full circles (like wheels). However due to the spacing between the
end
> points (on circumference) where the legs can come to rest on a surface
> (unlike wheels), they should be able to negotiate the stepped surfaces
> better.
>
> At any given time, two legs on one side (and motor) of the bot will
> remain inactive while the other side pivots over one leg and
advances.
> So the movement will be somewhat of a tumbling nature (may be we
should
> call this guy TUMBLER?).
>
> A further development can be tried to make the robot more adaptive to
> its environment. The bot starts with a given leg length (shortest -
> just enough to clear its underbelly thus requiring less torque and
power
> to move about). If it finds that it can not cross a step, it extends
> its legs (lehthens them - all 8 spokes at the same time for simplicity
> and balance). Then it again tries to cross the obstacle. The
> lengthening of legs could be a one step or multi-step process. It
> should be possible to achieve the lengthening/shortening of legs
through
> the use of a third motor or a stepper mechanism (a relay with a dog
> wheel/rachet?) for activation. Actual lengthening can be achieved by
> either a rack & pinion mechanism or with two concentric rings and
levers
> (think of the old Compur shutter mechanisms in some cameras- only
inside
> out).
>
> I have attached a very bad graphic representation of the tumbler & how
> to implement the leg extension. I hope the 4K gif file is not
> objectionable. The leg shape shown, allows one to choose between a
> rolling and a tumbling gait. The claw shape might help in getting a
> good foot-hold on steps (especially when augmented by ribbed rubber
> boots).
>
> While writing this, I started thinking. May be for a future
> competition, we can collect new ideas. Then base the competition on
> building better bots based (I like the sound of that :-) on one or
more
> chosen ideas. We can also have people collaborate so that people
(like
> me) who do not find the time to build can work with someone who does
or
> those with ideas needing special building skills/facilities can work
> with those who have access to such facilities.
>
> This is getting too long so I will stop. Thanks for reading. Let me
> know what you think.
>
> Dilip
> --
> Pl. remove *s from the e-mail address to reply
Home