Modified version of Bar’s Calibration simulation page. Apologies for the clunkyness, my programming skills have atrophied since retirement.
CalibrationLocal.html (41.2 KB)
1 Like
This is a really great improvement! Do you have any interest in making a pull request to make it the main version?
I have no problem with that except I don’t know how!
2 Likes
I will make a video! (although it might take me a little while) 
2 Likes
Another version, try the “Map Points” for a different view, see the points gradually coalesce until all points merge (in a perfect result which I’ve not seen yet). Also dumps 1st 8 values.
CalibrationLocal.html (51.8 KB)
Latest Version
FindAnchorsLocal.html (55.1 KB)
1 Like
could you make a page that lets you define the real anchor point locations,
Then have the ability to generate belt length samples for locations within the
work area
methods that occure to me off the top of my head:
- click on points and have it generate the data for those points
- be able to tell it to pick N points randomly within a AxB sized area
- be able to tell it to pick N points in a grid within a AxB sized area
The idea being to have a tool that generates belt length data that can be used
by various calibration simulators, including ones that use completely different
logic (which is why I’m requesting this as a stand-alone tool, not integrated
into one of your existing tools)
(future) enhancement(s):
be able to add in error factors on the reported belt lengths, such as
- belt stretch (reporting shorter than they should, I don’t know if this is a fixed percentage, or a percentage that increases based on length)
- angle related errors (arms hitting uprights, causing the belt lengths to
report shorter) - frame distortion (move the anchor points a bit randomly for each click)
- Z offset related errors (both individual anchors wrong, and systemic errors
like not accounting for wasteboard/workpiece thicknesses
David Lang