[Dprglist] Assisting PID odometry with an IMU
Murray Altheim
murray18 at altheim.com
Fri Jan 22 21:04:46 PST 2021
On 23/01/21 5:27 pm, David Anderson via DPRGlist wrote:
> Murray,
>
> Your PID controller should handle this. No IMU required. It sounds
> like the PID is not responding quickly enough when it first engages
> the carpet. Try kicking up the gains, or maybe just the D gain.
Hi David,
Hmm. I'd thought the tuning was okay in the sense that it seemed to
perform most of the time pretty well, with that bump onto the rug is
quite a steep one (from the wood floor it's a sudden 17-18mm jump for
a 70mm wheel).
As I've now got an RGB Potentiometer Breakout and a supporting Python
class on the robot it's relatively trivial to add a knob control to
the PID configuration:
https://shop.pimoroni.com/products/rgb-potentiometer-breakout
https://github.com/ifurusato/ros/blob/master/lib/ioe_pot.py
In looking at my PID configuration it looks like I'm only using P and D
right now:
kp: 0.09500
ki: 0.00000
kd: 0.00030
I'll certainly try some knob-tweaking to see how that makes a difference.
> From the circling in place video (bravo!) I think I see the robot's
> wheels speeding up and slowing down, like the controller is pumping
> or oscillating, rather than settling into a constant velocity. Seems
> like you still may need some PID adjustments.
I think I see what you mean, almost like the wheels are surging. I'll
try the test a few more times and also let it run at full speed for a
lot longer, then check the resulting logs to see if that's actually
happening or just some artifact of the video.
Thank you!
Murray
...........................................................................
Murray Altheim <murray18 at altheim dot com> = = ===
http://www.altheim.com/murray/ === ===
= = ===
In the evening
The rice leaves in the garden
Rustle in the autumn wind
That blows through my reed hut.
-- Minamoto no Tsunenobu
More information about the DPRGlist
mailing list