Alt-BEAM Archive
Message #07947
To: beam@sgiblab.sgi.com
From: BotDoc botdoc@yahoo.com
Date: Tue, 23 Nov 1999 14:24:54 -0800 (PST)
Subject: [alt-beam] Re: photowalker
As far as increasing function, adding a head probably
wouldn't improve much, if anything at all. As far as
the coolness factor goes, it would just be more
entertaining if it was able to look around as it
walks. It might be too difficult to get it to work as
well, but it is worth a try nonetheless.
justin
>My question is, why would you need a HEAD at all?
>I mean if it already goes to and locks on the
>brightest source of light.
>I'm sure i'm missing something here.
>_sparky
=====
A motor is a robot waiting to happen
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com
7948 Tue, 23 Nov 1999 23:46:34 +0100 [alt-beam] Re: photowalker beam@sgiblab.sgi.com Thomas Pilgaard Nielsen BotDoc wrote:
>
> Wilf,
> You are the man.
That is so true, !BUT! not to forget you (As of now I am attempting a
PCB design for your Bicore Photovore. It rocks!)
Anyway, would it be too much to ask if it is possible for you to post a
complete schematic on the phototropic bot. I am sorry to say that my
knowledge on electronics are rather limited - schemata would be a great
help.
Thanks in advance and cheers.
Thomas
> The schematic that you posted last
> night works like a peach. A lot of my hair thanks you
> as they get to hang out and party on my head a while
> longer. I put the circuit (minus the head motor) on
> my walker and it does display phototropic abilities.
> If the light is off to the side, it walks that way
> then straightens out when going directly to it (one of
> the led's starts blinking away when it is locked on...
> kind of helpful and cool to see). I'll put the head
> motor on later on today and start to fiddle with it in
> that way. Just had to make sure it could walk
> phototropically first (who said you can't make a
> phototropic 2 motor walker?). I am not quite sure how
> I am going to make the body keep up with the head,
> that could pose some sort of a problem as the head is
> going to be in the front of the bot. I thought that
> maybe some sort of one shot nv that diables the head
> motor when the body is swinging around to allow it to
> catch up might work but will work on it later. I'll
> let you know how the expirementing goes with it.
> Thanks again Wilf, you made my day... or week rather.
> Justin
>
> =====
> A motor is a robot waiting to happen
> __________________________________________________
> Do You Yahoo!?
> Bid and sell for free at http://auctions.yahoo.com
--
---
"Fremtidens computere vejer maaske ikke mere end halvandet ton."
Popular Mechanics 1949
7949 23-Nov-99,Time: 15:55:00 [alt-beam] Re: photowalker beam@sgiblab.sgi.com harsh@lanl.gov
7950 Tue, 23 Nov 1999 15:06:23 -0800 (PST) [alt-beam] will this work??? beam@sgiblab.sgi.com Evan Dudzik
--0-596516649-943398383=:1014
content-disposition: inline
I made up this little circuit, it needs some notes...
i had to make it in MS paint, so its not too good.
1st of all, the thing circled in red is NOT connected.
the things on the left are phototransistors from
inside an optocoupler, so the brain can run the servos
without screwups. the things in black are 555 timer
IC oscillators, which drive the things in grey, which
are hobby servos... BTW, there is only one servo, not
three, like show, the circuit just drives the servo
at different rates... heres how it is supposed to
work... the controller will bring either one or the
other phototransistor high (turning it on) thru the
optoisolator. this turns on the attached transistor,
turning on the next transistor, which allows current
to flow thru the oscillator #1, as well as the
servo... I know that the signal pin of the servo
should go to + and the ground should go to the output
of the oscillator. is this making sense at all?
Well, the first transistors after the phototransistors
are connected to a third transistor, and when either
of them turn on, they sink the current from the
oscillator and servo connected to that transistor, so
that the oscillator and the servo turn off... this all
has the effect that when the two inputs (via the
optoisolator) are both low, the oscillator to the
right goes to the servo, and the oscillator is such
that it centers the servo (supposing that anyone knows
how to operate a hobby servo). when either one is
high, that oscillator goes off, and the corresponding
one (to the input) turns on, making the servo move
(one oscillator makes it go 90 degrees left, the other
90 degrees right (again assuming someone understands
how to operate these servos)... this way, a robot
has basic control of its servos (for legs) as in left,
right, center... a few things i dont know about, are
1) what do the resistors connected to the transistors
do? i have seen them connected to other circuits like
so, and thus I put them in there... 2) WILL THIS
WORK??? PLEASE HELP ME OUT HERE!!! this whole thing is
just a little piece of a giant project, in which I
would like to make something like either genghis or
attila (both from MIT)
with 12 or so servos 6 legs, 2 servos per leg, plus
maybe more for some aimable vision system), and this
circuit would control ONE servo, and could fit two of
them on a 4X optoisolator IC, which would mean a
circuit that would be repeated for each leg, leading
to more simplicity, since I would be etching PCB's for
this thing... so just etch 6 boards, and i'm set.
after that, its just a matter of a few PICs and some
clean programming, mixed in with a staggering sum of
money... and I have the potential for one of the most
advanced robots ever built by one person (if not THE
most), especially someone without a big fancy
electronics degree and not even a full high school
education (yet... getting there!)... I know it sounds
like a crazy dream, but It could happen If this
circuit works (or if I can make a substitute)... thank
you all for your help in advance, I know its a big
strain, and means actually thinking a bit for most,
but it is very important to me, and I really need to
have someone help me fix this (if it doesnt work).. thanks...
=====
+------------------------+
|http://surf.to/photovore|
|Photovores online! |
|Evandude Dudzik |
+------------------------+
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com
--0-596516649-943398383=:1014
Attachment: schem.gif
--0-596516649-943398383=:1014--
7951 Tue, 23 Nov 1999 15:39:01 -0800 (PST) [alt-beam] kool, MIT solar robot! beam@sgiblab.sgi.com Evan Dudzik this beastie looks simple, but with kool stuff...
radio communications!?!? it does, however, have a
solar panel, and stores power in a 1F cap...
http://www.ai.mit.edu/projects/mars-rovers/rockettes.html
=====
+------------------------+
|http://surf.to/photovore|
|Photovores online! |
|Evandude Dudzik |
+------------------------+
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com
7952 Tue, 23 Nov 1999 15:42:41 -0800 [alt-beam] Re: photowalker "'beam@sgiblab.sgi.com'" Wilf Rigter Cool and very biological. It makes about as much sense to use a head on a
walker as it does on your own body. It provides direct measurement about the
angle between the light source and the centerline of the bot. It could allow
a walker to back up to the light. Since the head moves much faster than the
slow turning walker it can track a target independent of the motion of the
bot. Walkers jerk around a lot when moving and that jerky motion could
easily interfere with signals from fixed photosensors. Besides this is only
the start: for example a loud sound could initiate a quick head scan to see
if there is a bright light associated with the sound. Then there was the
idea of mounting a web cam on a walker but you'd have to mount it on a 2DOF
head to avoid total barforama.
Wilf Rigter mailto:wilf.rigter@powertech.bc.ca
tel: (604)590-7493
fax: (604)590-3411
> -----Original Message-----
> From: BotDoc [SMTP:botdoc@yahoo.com]
> Sent: Tuesday, November 23, 1999 2:25 PM
> To: beam@sgiblab.sgi.com
> Subject: Re: photowalker
>
> As far as increasing function, adding a head probably
> wouldn't improve much, if anything at all. As far as
> the coolness factor goes, it would just be more
> entertaining if it was able to look around as it
> walks. It might be too difficult to get it to work as
> well, but it is worth a try nonetheless.
> justin
>
> >My question is, why would you need a HEAD at all?
> >I mean if it already goes to and locks on the
> >brightest source of light.
> >I'm sure i'm missing something here.
> >_sparky
>
>
> =====
> A motor is a robot waiting to happen
> __________________________________________________
> Do You Yahoo!?
> Bid and sell for free at http://auctions.yahoo.com
7953 Tue, 23 Nov 1999 16:13:46 -0800 (PST) [alt-beam] Re: photowalker beam@sgiblab.sgi.com BotDoc
--0-719885386-943402426=:28982
content-disposition: inline
Thomas, here is the schematic. Hope it helps
justin
BotDoc wrote:
>
> Wilf,
> You are the man.
That is so true, !BUT! not to forget you (As of now I
am attempting a
PCB design for your Bicore Photovore. It rocks!)
Anyway, would it be too much to ask if it is possible
for you to post a
complete schematic on the phototropic bot. I am sorry
to say that my
knowledge on electronics are rather limited - schemata
would be a great
help.
Thanks in advance and cheers.
Thomas
> The schematic that you posted last
> night works like a peach. A lot of my hair thanks
you
> as they get to hang out and party on my head a while
> longer. I put the circuit (minus the head motor) on
> my walker and it does display phototropic abilities.
> If the light is off to the side, it walks that way
> then straightens out when going directly to it (one
of
> the led's starts blinking away when it is locked
on...
> kind of helpful and cool to see). I'll put the head
> motor on later on today and start to fiddle with it
in
> that way. Just had to make sure it could walk
> phototropically first (who said you can't make a
> phototropic 2 motor walker?). I am not quite sure
how
> I am going to make the body keep up with the head,
> that could pose some sort of a problem as the head
is
> going to be in the front of the bot. I thought that
> maybe some sort of one shot nv that diables the head
> motor when the body is swinging around to allow it
to
> catch up might work but will work on it later. I'll
> let you know how the expirementing goes with it.
> Thanks again Wilf, you made my day... or week
rather.
> Justin
>
> =====
> A motor is a robot waiting to happen
> __________________________________________________
> Do You Yahoo!?
> Bid and sell for free at http://auctions.yahoo.com
--
---
"Fremtidens computere vejer maaske ikke mere end
halvandet ton."
Popular Mechanics 1949
=====
A motor is a robot waiting to happen
__________________________________________________
Do You Yahoo!?
Bid and sell for free at http://auctions.yahoo.com
--0-719885386-943402426=:28982
Attachment: photowalker.gif
--0-719885386-943402426=:28982--
7954 Wed, 24 Nov 1999 18:45:26 -0600 [alt-beam] Re: photowalker beam@sgiblab.sgi.com SG At 03:42 PM 11/23/99 -0800, you wrote:
>barforama.
>
> Wilf Rigter
Heh HA!
yup, a head would allow independant survelance apart from the body...
besides, the coolness factor. you know?
Home