ASTAP as plate solver

Suggest to add ASTAP as a plate solver. It can be executed using a command line in either PlateSolve2 style or its own native command line style. One of the advantages is that it can do blind solving similar as Astrometry.net.

Han

Would be very cool if you could integrate this with Voyager’s Application Server so you could load, view, analyze and potentially stack images as they are acquired - there’s a “new FITS ready” signal from Voyager to trigger the load.

https://voyager.tourstar.net/index.php?title=Application_Server_API

FYI, I got a popup with an “Invalid floating point operation” exception when loading a FITS image using your latest Win64 build.

Cheers,
Rowland

I haven’t build in a client or server in ASTAP yet. That could happen in the future. I haven’t figured out either how users are suppose to view FITS files produced by Voyager.

For the moment, I’m only suggesting to Leonardo to add ASTAP as solver. The simplest implementation would be to use it in PlateSolve2 mode where command line and output file are compatible.That would be a very small change.

About your “Invalid floating point operation” while loading a FITS file. If this is a repeatable error, could you send me that file/post somewhere for analysis?

Han

Hi Han,

Here’s a link to the file which causes the error when opened in ASTAP:

As far as viewing FITS with Voyager, there are several options:

  1. Use the free Deep Sky Stacker Live and monitor the folder where Voyager is saving images. New images will show up automatically in DSS Live
  2. If you use Maxim DL or TheSkyX for your camera, they both contain viewers that show images as they are taken
  3. A local astro imaging friend, mikefulb on this site, built a viewer in Python that integrates with Voyager the way I suggested you could do with ASTAP. It is open sourced and Mike posted a link on this forum to retrieve it
  4. And finally, just use any FITS viewer and open the files manually. I agree however, it is nice to use an option that shows the images automatically as they are taken.

Leo has commented here that a viewer is on his to-do list, and that will certainly satisfy a lot of people’s desire to have something that is bundled with Voyager. I think once you become used to Voyager’s reliability, the need to monitor your images as they are taken declines compared to other software, but I agree, it is a very common request and we are visual creatures who like to see what’s happening!

Cheers,
Rowland

Hi Rowland,

The file you provided loads normally. Maybe it was open in an other application? If so i should improve the message. Anyhow even while very noisy it solves with default settings:

  • 14:41:57 Creating monochromatic x 2 binning image to solve.
  • 14:41:59 338 stars selected and 256 tetrahedrons selected in the image. 319 database stars required for the square search field of 0.9°. Overlap set at 0% based on the number of tetrahedrons.
  • 14:41:59 Using star database g17
  • 14:41:59 91 of 99 tetrahedrons selected matching within 0.005 tolerance. Solution x:=-0.666729x+ -0.630009y+ 2660.298601, y:=-0.630210x+ 0.666495y+ 293.433587
  • 14:41:59 Solution found: 09:55 34.2 +69d03 56 Solved in 2.9 sec. Offset was 0.000 deg. Used stars up to magnitude: 16.2

Since it is pretty large, set in ASTAP downsample to 2 and save settings. In downsample 1x it takes more time and is less reliable.

Attached with annotation.

ASTAP has a FITS thumbnail viewer. It is slow and doesn’t rescan automatically but that could be reviewed.

Cheers, Han

Hi Han,

Thanks for checking it out. The image did open and display, but I kept getting the “invalid floating point” exceptions. I just tried again and it’s working now, so maybe it was open. It wasn’t loaded in any viewer but it had been moved to this PC from another one by a file explorer application.

I shut everything down and tried again and it worked OK this time.

Computers :slight_smile:

Thanks again and cheers,
Rowland

Hi Han, is on to-do for sure … i’m finishing API documents for 10 Micron software and model alignment creation and the FIT Viewer for Voyager … after i can schedule the ASTAP interface

All the best
LO

I have the impression that communicating via a server makes it too complicated. My first impression it would be easier if Voyager could execute a viewer (path configurable) with the fits filename as parameter every time a new file is created. If you do this with ASTAP, it will create a new instance every time. I’m just experimenting with the possibility to restrict this to one. So if you click on fits filename, they are opened in the same ASTAP. Experimental. You could download this version as:
http://www.hnsky.org/astapx.zip

If you want to use the solver and viewer at the same time, you have to duplicate the executable and rename one.

Han

Wow Han … this is a great job. Congrats … a lot of job.

I can add possibility to open FIT in ASTAP, tab windows option can be a good things with max tab number before delete old FIT

If you implements client server you can do stack in real time, Voyager can send image fit name to stack

Today I improved the concept. The ASTAP “one instance only” limitation is only maintained if the program is executed with one command line parameter and the parameter is not a platesolve2 command. (assuming the platesolve2 command ends with a number) So you can start the astap.exe twice, three times as you like. However if you associate ASTAP with fits files and you click on a fits file, ASTAP is executed with the file as parameter and therefore opened in the first and only instance of ASTAP. So Voyager could execute the following commands and only one ASTAP instance will handle them :

c:\programs files\astap\astap.exe file1.fit
c:\programs files\astap\astap.exe file2.fit
c:\programs files\astap\astap.exe file3.fit

I assume this viewer mode is useful if new camera files are available. No need for server link. I also have to keep ASTAP universal and that’s the case with this setup.

At the same time an ASTAP can be executed in parallel for solving. So one in viewer mode and the second in command line mode to solve in parallel.

Download this version 0.9.208aX2 from the same link as in my previous post. In a few days I will release this in the normal way as 0.9.209.

>> tab windows option can be a good things with max tab number before delete old FIT

The viewer doesn’t have tabs. Introducing that looks with all mouse interactions a little complicated and the same could be realized by more instances. More instances is now less possible.

>>If you implements client server you can do stack in real time, Voyager can send image fit name to stack

For the moment, I’m not enthusiastic about live stacking. I prefer to do that after data collection phase, so the next day.

Last thing, implementing an internal viewer in Voyager can’t be too complicated. That would be the best option.

Thanks Han,

i have written all on my todo list , i’ll add an automatic way to open ASTAP like viewer.
Thanks a lot for your work.

Leo

I would like to switch over to Voyager, but i’ll wait for ASTAP support, any idea when that will be added?

1 Like

Hi
Check out the process which is described on the website of Han:
http://www.hnsky.org/astap.htm#sgp
under “voyager”
Ed

1 Like

We will add native support for ASTAP soon.

All the best
Leonardo Orazi

2 Likes

I just tried this and could solve an image loaded into OnTheFly. I got an error box: “Please Connect Plate Solve Control”…kinda cryptic, not sure what to do next

Try to read this section of wiki if message is cryptic, can help https://voyager.tourstar.net/index.php?title=Startup

Thanks, guess I should’ve RTFM. Just started the demo and really liked ASTAP on SGP and jumped right into trying to get it to work with Voyager without reading further. The information on hnsky.org is correct and ASTAP works great!

No problems, you ar welcome. Some of the controls in Voyager are external program that need connection. For this reason all the control are available under Voyager only after connection. Voyager is a system integrator and really different from SGP.

All the best
LO

Blind solving is really rare … probably just the first time if you are not in fixed observatory environment.
So timing on solving I dont think is mandatory … anyway we will adding ASTAP for sure.

All the best
LO