Searching \ for '[OT] Best way to get started with robotics' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=best+way+get+started
Search entire site for: 'Best way to get started with robotics'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Best way to get started with robotics'
2006\02\28@081719 by Bill Kuncicky

picon face
I'd like to begin doing some experimenting with robotics, using a pic.  
Could I get some suggestions on how best to start?  I've used Google to
look up some websites, but a lot of  them seem to use Basic Stamp
stuff.  Which is OK, except that I want this to be part of my learning
how to program pics.  A lot of places also seem to seem to sell very
expensive "robot kits" and I'm pretty much on a low budget.  I found a
used book called "Robot Building for Dummies" by Roger Arrick, President
of Arrick Robotics, and have looked through it.  But it seems to push
primarily the stuff that he sells.  Any help would be appreciated.  My
budget for getting started would probably be around $100, but I already
have the pic part (chips, a Wisp628, test equipment).

Bill

2006\02\28@084017 by James Humes

picon face
Hi Bill,

    All you need to do is add motors and sensors to your PIC and shake
well.  I think the hardest part starting out is getting all the mechanical
parts connected solidly.  http://www.lynxmotion.com offers gearhead motors and hubs
that match their shafts.  This helps.  Use an H-bridge IC to drive them.
Add a couple of sensors and you have the makings of a robot.    Things like
wheel encoders (for dead reckoning) or sonar would be nice next steps.

James


On 2/28/06, Bill Kuncicky <spam_OUTkuncickyTakeThisOuTspamgmail.com> wrote:
{Quote hidden}

> -

2006\02\28@091602 by Jason

flavicon
face
If you haven't found it yet, visit http://www.robotroom.com

Consider looking into a cheap (~$10) remote control car.  Just don't bother
trying to reverse engineer the controller board.  Build your own to directly
connect to the motors through an H-Bridge.

I bought one of those sumo robot kits for $100
http://www.amazon.com/gp/product/0071411933/ref=pd_sbs_b_2/103-7897699-1072668?%5Fencoding=UTF8&v=glance&n=283155.

It was a big mistake for experimenting with robots.  It's an interesting
toy, but you have to go through it's own basic stamp programming that's just
a pain and not very useful for learning practial robotics stuff imo.



From: "Bill Kuncicky" <.....kuncickyKILLspamspam@spam@gmail.com>
Sent: Tuesday, February 28, 2006 5:17 AM


{Quote hidden}

2006\02\28@094524 by kravnus wolf

picon face
www.myke.com/prc-book.htm
a book i own and recommend to you. able to get you
to build robots fairly quickly but the problem as with
all robot projects is the parts. ranging from
motor to sensors. for rapid prototyping consider
mindstorm. But if you want to build multiple bots then
you left with one choice. Build one from scratch.

Good luck,
John

--- James Humes <james.humesspamKILLspamgmail.com> wrote:

{Quote hidden}

2006\02\28@100145 by Jamie C. Pole

flavicon
face

Hi Bill...

Parallax sells a hardware kit that includes all of the Boe-Bot parts  
except the Basic Stamp and controller board.  The part number is  
28124, and the current cost is $109.00.  This kit also includes  
servos, but no sensors (except the "whiskers").

I can understand why people don't like the SumoBot for learning, but  
the Boe-Bot is really a good platform.

Jamie


On Feb 28, 2006, at 8:17 AM, Bill Kuncicky wrote:

{Quote hidden}

> --

2006\02\28@101701 by sergio masci

flavicon
face


On Tue, 28 Feb 2006, Bill Kuncicky wrote:

{Quote hidden}

The biggest problem I found was mounting the servos to give strong well
articulated legs. This is my (unfinished) effort

http://www.xcprod.com/ROBOT/

Driving the servos is easy. If you follow the XCSB links you will find a
free driver for 4 (simultanious) servos using a 16F628. You can easily
control 20 using a 16F876 and the servo library (not free btw - but
cheaper than bying a dedicated controller board).

Regards
Sergio Masci

http://www.xcprod.com/titan/XCSB - optimising PIC compiler
FREE for personal non-commercial use



.

2006\02\28@112505 by D. Jay Newman

flavicon
face
> I'd like to begin doing some experimenting with robotics, using a pic.  

For information I would suggest the usenet group comp.misc.robotics,
and the Yahoo group "theroboticsclub".

As to PIC robotics, that is how I got my start. I would get a basic
base with modified R/C servos because you don't need anything else
to control them.

I made a robot once with such a base, a breaadboard, a PIC (and the
few passives needed to make it work), a battery, and some wires.

The base I used came from Budget Robotics (http://www.budgetrobotics.com/).

Good luck!
--
D. Jay Newman           ! Author of:
EraseMEjayspam_OUTspamTakeThisOuTsprucegrove.com     ! _Linux Robotics: Building Smarter Robots_
http://enerd.ws/robots/ ! (Now I can get back to building robots.)

2006\02\28@114339 by David VanHorn

picon face
Why bother with someone else's servo code?

The basics aren't that difficult.
I did an 8 channel driver in the AVR, using a single timer to manage all 8
servos, with individual scaling of min/max pulse width so that I could mix
and match servo types and still get the same range of motion for the same
position input.  There is some difference between brands, more than I
expected when I started the project.

I stopped at 8 because it was a nice round number, and I didn't really have
any specific need, it was just a fun project.  It would be easy to expand to
16 or more, and at those scales, you wouldn't need to generate the frame
delay, you could just round-robin the servos and let that do the frame delay
for you.

That demo software was generating morse code on an I/O pin, scrolling two
displays, one LCD and one VFD (very different interfaces), managing the 8
servos, and full duplex 232, as well as a stepper output, and some other
fun, and I didn't even come close to maxing out the chip at 8 MHz clock.

2006\02\28@124634 by Eisermann, Phil [Ridg/CO]

flavicon
face
piclist-bounces@mit.edu wrote:
> Why bother with someone else's servo code?
>
> The basics aren't that difficult.
> I did an 8 channel driver in the AVR, using a single timer to
> manage all 8 servos, with individual scaling of min/max pulse
> width so that I could mix and match servo types and still get
> the same range of motion for the same position input.  There
> is some difference between brands, more than I expected when I
> started the project.

Especially true if you intend for it to be a learning project.
You can spend a few minutes googling to get all the info you
need on controlling servos. Basically its just a pulse of a few
milliseconds, repeated every 20 milliseconds.

I also rolled my own, except for a specific application. I wanted
to be able to test the range of motion on an RC sailplane I was
building without having to charge up the transmitter all the time.
As Dave says, there enough difference among brands that you want
to take the min/max range into account when you start.

2006\02\28@130936 by Lindy Mayfield

flavicon
face
This is a very interesting position to be in.  

To be up front, I'm a long time lurker on Piclist and a robot builder wannabe.  So my opinion isn't worth that much, but I've thought about this a LOT.

But here is what I think.  You can attack robot building from many angles.  There are the mechanics which I suck at.  This means getting all the pieces parts to work together and be controlled somehow, either by direct or remote control or by a controller.

When the mechanics are working properly you can go further with microcontrollers.  But you can only control what you can "see", so your robot needs sensors and feedback.  

My experience after building Arrick's robot was that it was interesting, but a let down.  And the support wasn't that good.  I screwed up something by clipping a part incorrectly because of an error in the book.  I called and they sent me a replacement part which was damaged in the mail.  But after repeated calls and emails I never got a second replacement.  And that simple part I couldn't find here in Europe.  )-:  

But it wasn't so much the support that was a let down, as my expectations.  To do what I wanted to do I needed a lot of sensors.  I think this is normal for most people, though, when they first play with robots.

There are many different types of robots you can experiment with, like BEAM or arms or humanoid, etc.

But if you are interseted in the software aspects there are other choices.  If you know Java (or want to learn it) you can concentrate on the software and AI and create robots that fight each other.  http://robocode.sourceforge.net/

You can do in Java without worrying about the mechanics of the robot much more than you can do with a Pic.

But Pics are absolutely facinating, too, and there are tons of projects you can do with them if you want to learn without getting so bogged down with the robot mechanics.  I mean, what will our robot do?  Go forward until a sensor detects an object.  Then turn, and try again, etc.  Believe me that gets boring quick. (-:

Anyway, that is just my experince.  I don't mean to discourage you with robots.  But the Pics and the things they can do are way, way cool.

Lindy





{Original Message removed}

2006\02\28@131510 by David VanHorn

picon face
The AVR code I wrote is on dontronics, the GSV5 codeset.
It's a whole bunch of routines, all wrapped up in a cooperative
multitasker.

It's been a while since I worked with it, but you can just trim out the
parts you don't need.   I just wish there was a more elegant way to handle
the dependencies between different files.

2006\02\28@171837 by Wouter van Ooijen

face picon face
> My
> budget for getting started would probably be around $100, but
> I already
> have the pic part (chips, a Wisp628, test equipment).

If you have chips and a progger the next items are motors, sensors and
an accu. Steppers are a cheap but low-power choice, motors with gears
are nice but can be expensive, hacked servo's seem to be often used.
Check for instance my site for mot-11, surplus stores like
electronic-goldmine, for motors. Sensors can be switches, IR,
ultrasonic, or even a low-cost LED to follow a line.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


More... (looser matching)
- Last day of these posts
- In 2006 , 2007 only
- Today
- New search...