Subject:
|
PID controller HELP!!!!!
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Thu, 20 May 1999 12:56:23 GMT
|
Original-From:
|
BOB_KELLY/FIRST_UNION@FIRSTUNION-REITstopspammers.COM
|
Viewed:
|
948 times
|
| |
|
|
Hello all....
I am currently trying to implement some level of shaft encoding/motor
control on the handy board. I am using the shaft encoder software available
on the H board site. I am currently using the "slow" version, as the top
velocity reading I have been able to get is a 13 (same as if I used "fast"
version). I have 4" wheels and am using reflectance encoders on a
alternating black and white disk with 32 black stripes per revolution. With
a velocity of 13 (motors full on) my actual speed is much faster than is
needed (and safe!) for my project.
I have created a simple software driver that will allow me to control both
implemented motors with the following go(DES_POWER_LEVEL,
DES_DIRECTIONAL_BIAS). All that is needed to "slave" the wheels at this
point is a reliable encoder velocity for each wheel that can then be
inserted into my go() program.
The problem is, the velocity measurement appears to be too coarse ((int)13
steps) and does'nt provide reasonble resolution. Is there a way to change
the output value of the shaft encoding software to a "float" type of value?
Ideally, the velocity count would max at about 90% of full motor power. I
know I could use a conversion factor, and I would if the encoder readings
were not so rough. Should I create new encoder disks with two or three
times the resolution? I am concerned that at this point that I would exceed
the mechanical limitations of the Hammamatsu detectors (as used in "Mobile
Robots..." book) or the ability of the software encoder counter to keep up
(even with the fast version). Any thoughts on this?
Also, my problem may be the steps of the existing motor control driver for
the Handy board. I have loaded the Pcode (or is it the ICB?) that makes
the LCD print "Smooth PWM" on the start up screen and I also have the
latest Expansion (3/19/99) board handyboard libraries, but it still appears
to be adjusting the motor control in rather large steps.
Please help, this is extremely frustrating (but still kind of fun in a
masochistic kind of way)...
Thank you all,
Bob Kelly
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
