Autofocus improvements - Githubissues

AnthonyHorton commented 6 years ago

Experience during Huntsman commissioning shows that while the autofocus routine is working well there are still a few improvements that would help to make it more robust, paving the way for remote observing.

[x] Automatically run a coarse focus prior to the first observations of the night, then run a fine focus prior to each block of observations (latter already happens).
[ ] If best focus appears to be outside the range of the focus sweep automatically extend the range with additional focus images and try again.
[ ] Determine whether there are stars in the image and if not (e.g. vignetted by dome, clouds) discard the results and issue a warning.
[ ] Report whether focus failed (e.g. no stars) and, if successful, some measure of confidence in the results.
[ ] Add information about most recent successful focus (time, initial, final, confidence) to FITS headers
[ ] Add an option to follow autofocus by deliberately defocusing by a set amount, for bright star photometry.
[ ] Focus prediction based on ambient temperature and empirical temperature - focus relation (longer term)
[ ] consider discussion here: https://aao-org.slack.com/archives/GQ9F8Q5FY/p1602583198298800
[ ] change focus setting based upon current filter https://aao-org.slack.com/archives/GKNFB1TEH/p1611806150375900

AnthonyHorton commented 6 years ago

@wtgee Will be picking your brains about how best to implement the first point.

lspitler commented 6 years ago

address focuser issues

AnthonyHorton commented 6 years ago

Implementing the first item with https://github.com/panoptes/POCS/pull/659 and #75

lspitler commented 4 years ago

AstroHuntsman / huntsman-pocs

Autofocus improvements #69