SW&IT Meeting Minutes 2018-07-23
Attendees: Andrew, David, Doug, Igor, Kellee, Matthieu, Petr, Stephen, Xianyu, Yang
News from management
Previous Action Items
Bring any topics you want for discussion at the meeting here. Use <topic>[<name>] formatting.
This week in github
- Tagged VLAN on jet computer (SH)
- Final draft of handover document
- Closed out remaining handover items
- Tried testing swing arm script but there was a network glitch (now resolved)
- Test swing arm script
- Test interface with TCS GUI
- Progress on IT email lists.
- Work with Robert's intern re: getting files into the vault from the OSU MODS laptop
- Synology DSM update done, IT 6691 closed
- Get double-confirmation on member list for Instruments.
- Work on power plan refresh
- Create directory for external storage of Vault backups (currently stored on the same virtual disk as the VM)
I spent last week working on LUCI
I created a python script that mimics the LUCI
- double correlation (subtraction of two cube images, where each frame on the cube represents a subwindow on the detector). The subwindow is tipically centered on a square pinhole)
- three pass sigma clipping (an algorithm that recursively applies the standard deviation on a column of pixels from the cube and weeds out the extreme values (hot pixels and anomalies)
- background shifting zeroing negative numbers (this is basically to remove the bias on the image gausian and bring it to zero)
- momentum centroid calculation (a simple algorithm that calculates the center of image based on intensity/weigth)
to that logic we added two modification to two of the above algorithms:
- The three pass sigma clipping gets replaced by the median
- the background shifting zeroing negative numbers gets replace by a square donut background mean, where everything is biased to zero, no zeroing of negative numbers.
With the script we hope to understand a bit more the characteristics of the data. The current LUCI
logic seems to work well in most cases, except when we hit specially bad row of pixels or the subwindows are not that well aligned.
So there is some work to be done yet. We try to get as much done before LUCI2 is taken off the telescope.
- Load new set of AFC coordinates to LUCI database.
- install the AFC python script somewhere for someone (hopefully) to use it.
- Vacation from the 24th to the 29th, but will be available if someone needs me.
- Put LBC, MODS, LUCI-1, AO-DX, AO-SX into "maintenance" on Status Server for the Summer.
- Took LBC, MODS, LUCI-1, AO-DX, AO-SX out of Alarm Handler for the Summer.
- Pull request completed for OAC updates to GCS in 2018A. Created a new GCS patch for the mountain on tcs-test.
- Pull request completed for a GCS AzCam update I had in SVN trunk. I have relinquished GCS to Petr.
- OAC 5.0 is installed on the mountain, but not released yet. Started digging in to how OAC can get the UMAC data similar to what OSS gets into telemetry. The AGW UMACs are harder to interface to, so it may not be possible.
- Vacation on Thursday.
- Vacation Monday and Tuesday.
- I'll be in on Thursday morning - cleanup OVMS update/notes; OAC updates - Dan has some issues.
- M3 OSS control on laptop (installed in virtual box, proved to work, remote connection to laptop available for help)
- OVMS kernel update
- final touches on Time To Live (added events when less than 30m and 1m left, monocular display, do not show TTL when PCS is not controlling the mount)
- fixed PMC bug (DUMMY mirror when PMC was started before VxWorks controller)
- finish OVMS update (new SSD)
- collimation work
- ARGOS arm control through MCSPU/OSS.
- Replace firewalls on Friday
- Small meetings to clean up R3 in CRA
- Prepare for DNS change sometime in September.
- Opened InformationTechnology/UpgradeOBS.
- Python upgrade for Shane.
- Support summer shutdown activities, i.e, adsec monitor script, ALH for AO;
- Support restart SX AdSec restart for LBTI-SOUL test;
- AdSec actuator jumping discussion;
- Read the MG electrinics design report.
- finish read the MG electrinics design report;
- check the AOS code in git;
- support some other summer shutdown activities.
- Working on IT issue #7325 to develop a tool for mining the collimation data.
- A preliminary version is completed.
- Working on some performance improvements as well as new features.
- Should be able to finish it soon, and after that, will study collimation data more closely with Petr.
- Working on pointing model documents.
- Continue to study the AO supervisor code and the IT issue #7261 (memory problem with WfsArbitrator)