August 28, 2006, 1:07 am
Hello all. I have been working on a robot project with a group of
friends and we have encountered a slight problem in making the robot
move properly. We decided to use stepper motors as the drive train
system. Using stepper motors we know the relative telemetry of the
robot but we have no way of compensating for the stepper motor skipping
a step or wheel slippage. While debating different solutions we came up
with the idea of using two optical mice mounted under the robot facing
the ground to track ground movement as a secondary telemetry for
correction.
I would like to know if anyone in this group has done this set up
before and can post suggestions.
Thanks
Andy
friends and we have encountered a slight problem in making the robot
move properly. We decided to use stepper motors as the drive train
system. Using stepper motors we know the relative telemetry of the
robot but we have no way of compensating for the stepper motor skipping
a step or wheel slippage. While debating different solutions we came up
with the idea of using two optical mice mounted under the robot facing
the ground to track ground movement as a secondary telemetry for
correction.
I would like to know if anyone in this group has done this set up
before and can post suggestions.
Thanks
Andy
Re: Opitical Mouse for tracking robotic movement
Someone posted a URL with a picture of a bot they were helping high school
kids build in the past few months. It used an optical mouse. But instead
of using it to track ground movement, they used it to track the wheel
movement as a cheap shaft encoder. The wheel was large enough that they
simply placed a sheet of paper against the wheel and mounted the mouse
against the wheel looking at the paper (if I am remembering correctly)). I
don't remember the name to Google for. Maybe the guy that posted will read
your post and answer or maybe someone else here will remember who it was?
I don't know how accurate the mice are however. They don't need to be
accurate to work as a mouse since you have constant feedback with the user.
They might have more error than your stepper and wheel slipping combined.
--
Curt Welch http://CurtWelch.Com/
curt@kcwc.com http://NewsReader.Com/
Re: Opitical Mouse for tracking robotic movement
Ah, I was able to find it with a bit of google work.
Here's a video news report about the bot ...
http://video.google.com/videoplay?docid 65197422869440629
And here's an article about it:
http://www.linuxjournal.com/article/9103
Both talk about how they used the optical mouse.
--
Curt Welch http://CurtWelch.Com/
curt@kcwc.com http://NewsReader.Com/
Re: Opitical Mouse for tracking robotic movement
The above project used a standard PC board running Linux which means it
already had support for a standard mouse. It was probably easier for them
to use that than to interface some type of encoder.
But sure, in general, shaft encoders are easy and cheap to build and would
probably be more accurate than the mouse.
--
Curt Welch http://CurtWelch.Com/
curt@kcwc.com http://NewsReader.Com/
Re: Opitical Mouse for tracking robotic movement
Agilent (HP) makes the sensors (i don't know of other vendors) and you can
find data sheets here;
http://www.avagotech.com/products/document-results.jsp?catIdI39
one example is;
http://www.avagotech.com/assets/downloadDocument.do?idA5
some points to note - maximum speeds top out at about 14 inches/sec, and if
sensing ground, depth of focus is typ 0.5mm.
Site Timeline
- » DC MOTOR with 150:1 gear ratio on Ebay
- — Next thread in » General Robotics Forum
-

- » geeks.com specialing out "boe-bot"
- — Previous thread in » General Robotics Forum
-

- » evoMUSART 2013: First CFP (with correct dates)
- — Newest thread in » General Robotics Forum
-

- » DCC sound question
- — The site's Newest Thread. Posted in » Model Railroad Forum
-


Subject







