Applying the Boe-Bot Digital Wheel Encoder Kit (28107) by Philip C. Pilgram

kevinb2050

Has any one used the Parallax article Applying the Boe-Bot Digital Wheel Encoder Kit (28107) by Philip C. Pilgram. My question is that I am running my·heavily modified boe-bot (wider wheel base) and with a parallax USB servo controller and was wondering if these mods would affect the results of this article.

Phil Pilgrim (PhiPi)

The wider wheel base will affect calibration. So long as it's not so wide as to cause overflow in the calibration program, it shouldn't be a problem. If you're using an external servo controller for the wheel motors, you should still be able to perform wheel odometry. Coordinated motion will elude your efforts, though, since the Stamp program needs to be able to control individual servo pulses to keep the motors tracking with each other.

-Phil

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
'Still some PropSTICK Kit bare PCBs left!

kevinb2050

Talk about going direct to the author, hi Phil
the wheel base is about 1 1/2" wider to allow for the extra boards, twin battery packs etc.
I have added the Servo controller as I ran out of I/O which killed my motion control program based on time/measurements as I could not determine pulses as you have pointed out. I thought that by using the encoders i could shut down the servo's after x encoder pulses thereby getting coordinated motion.

kevinb2050

Sorry another question should I run the calibration program with the servo's connected to a Board of education board and not thru the servo controller

Phil Pilgrim (PhiPi)

No, run the calibration program with the controllers you intend to use.

We may have different ideas about what "coordinated motion" means. When I use the term, I mean "a constant speed ratio between two wheels, even if one of them gets slowed by external means". For that, precise control of every servo pulse is essential. If you just want to count encoder pulses and stop, your servo controllers probably won't get in the way.

-Phil

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
'Still some PropSTICK Kit bare PCBs left!

kevinb2050

All i need to achieve is
1 a. Bot wanders under own control
b. Bot gets remote command to return to it's starting point
2. Bot is given x/y coordinates to go to from known start position
and hopefully
3. Bot roams randomly then recieves command to go to x/y coordinates
I am not looking for pin point Accuracy but for it to be in the general area i.e 6 to 12 " of x/y
Am I on the right track with the encoders to establish a programable GPS system, as I can not think of an alternative, as the bot is for indoors a GPS unit is not possible.
Thanks again for your time

Phil Pilgrim (PhiPi)

It all depends on how much wandering the bot does before being commanded to a certain position, how far away that position is, the kind of tires you're using, and the kind of surface you're rolling on. Even under the most ideal circumstances, any wheel odometry system will accumulate errors and need to coordinate with known external reference points from time to time. Not even the Mars Rovers, which rely on wheel odometry extensively, are immune to losing their bearings without the occasional sun shots, comparisons with visible landmarks, and corrections from orbiting satellites. But the wheel encoders will certainly give you a huge advantage over navigating blindly, especially if your area of coverage is relatively confined and your floor is smooth.

-Phil

kevinb2050

Thanks for your help - have ordered my encoders should arrive in Australia in about a week - then time to play - will let you know how i get on - thanks again for your time
Kevin