[Dprglist] From DPRG RBNV last night - about that heading loop glitch when rotate CW from 360 thru 0

[Carl Ott]

As discussed last night.

Karim was right - it was easy.

Plus - we now have *additional proof that algebra doesn't work on the
weekends when the weather is nice outside.*

instead of cancelling out to zero, this term somehow wound up getting
doubled
[image: image.png]

the ugly logic below still wants to be simplified & cleaned up.  But at
least it now works correctly:

the method below calculates *fbTerm* for this summing node

*setpoint - fbTerm = loop error *

where *setpoint *has range [ 0 .. 360 ]
where* IMU outputs* [ 0 .. 360 ]
and where *error *has range [ -180 .. +180 ], to give the shortest rotation
from current to setpoint

[image: image.png]

Anyhow...
Now onto cleanup, then to actually tuning the loop, then finally to more
interesting behaviors...

Carl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.dprg.org/pipermail/dprglist-dprg.org/attachments/20210324/d4f750d0/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 19837 bytes
Desc: not available
URL: <http://lists.dprg.org/pipermail/dprglist-dprg.org/attachments/20210324/d4f750d0/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 58733 bytes
Desc: not available
URL: <http://lists.dprg.org/pipermail/dprglist-dprg.org/attachments/20210324/d4f750d0/attachment-0003.png>