Realigning grids

[Shtantan1]

Background:

I'm attempting to create a small catalogue of star maps/planispheres from the perspective of observers on nearby celestial objects as a little demonstration of just how far away the stars are, what stays the same/what is different, etc.

This proved easy with objects in the solar system using Stellarium, which allows for a 180 degree FOV. Stellarium doesn't let you fly out of the greater solar system, though, so I've turned to Celestia.

I've attempted to force Celestia's FOV up to 180 by modifying URLs and have been successful in doing so... but the result if unfortunately glitchy and is not usable for my project. With FOV not being the solution, my best bet appears to manually stitch screenshots together in Photoshop. I've the image manipulation ability but could really use reference guides.

Question:

I've noticed that the wireframes for all of the stars that Celestia renders are oriented in the same way as the sun, lining up with the ecliptic grid. Plug-in stars and planets I have are seemingly all oriented the same way. The orbital plane of binary/multiple stars, though, such as Alpha Centauri, Sirius, 61 Cygni, etc seem to have inclination information built into the program.

Let's say this: I would like to align a new grid or realign an existing grid with the orbital planes of nearby binary stars. Alpha Centauri A and Alpha Centauri B.

I take it this is not possible through the user interface, but could it be done by manipulating a line of code somewhere? Like "permanently" altering the orientation?

I've searched for plug-ins but I'm unsure of how exactly a question like this should be worded.

[selden]

by Shtantan1 » 03.01.2018, 14:42
Background:

I'm attempting to create a small catalogue of star maps/planispheres from the perspective of observers on nearby celestial objects as a little demonstration of just how far away the stars are, what stays the same/what is different, etc.

This proved easy with objects in the solar system using Stellarium, which allows for a 180 degree FOV. Stellarium doesn't let you fly out of the greater solar system, though, so I've turned to Celestia.

I've attempted to force Celestia's FOV up to 180 by modifying URLs and have been successful in doing so... but the result if unfortunately glitchy and is not usable for my project. With FOV not being the solution, my best bet appears to manually stitch screenshots together in Photoshop. I've the image manipulation ability but could really use reference guides.

Imagemagick or MMPS could be used to assemble framegrabs into a single image with an appropriate projection. They're command-line image manipulation programs.

Question:

I've noticed that the wireframes for all of the stars that Celestia renders are oriented in the same way as the sun, lining up with the ecliptic grid. Plug-in stars and planets I have are seemingly all oriented the same way. The orbital plane of binary/multiple stars, though, such as Alpha Centauri, Sirius, 61 Cygni, etc seem to have inclination information built into the program.

Most of Celestia's stars are defined in the binary file stars.dat. It
was derived from the Hipparcos catalog which contains no orientation
or orbital information.

Stars which have orientation and orbital information within Celestia
have to be defined in textual .STC (STar Catalog) files. Stars known
to be within 25LY of us are defined in nearstars.stc, including Alpha
Centauri and our Sun. (A quick look in that file found only three
stars with orientation inclinations specified: our Sun, Vega and
Fomalhaut.) Some binary star systems have their orbits defined in
spectbins.stc and visualbins.stc.

The structure of .stc catalogs is described in
https://en.wikibooks.org/wiki/Celestia/STC_File

Let's say this: I would like to align a new grid or realign an existing grid with the orbital planes of nearby binary stars. Alpha Centauri A and Alpha Centauri B.

I take it this is not possible through the user interface, but could it be done by manipulating a line of code somewhere? Like "permanently" altering the orientation?

I've searched for plug-ins but I'm unsure of how exactly a question like this should be worded.

One way to show a grid would be to use an all-sky .png or .jpg image
with appropriate lines and labels.

One way to show an all-sky image in Celestia is to define a very large
spherical object using a .DSC (Deep Space Catalog) file centered on an
appropriate xyz location and tilted appropriately. An appropriate
image can be projected onto the inside of that sphere. That image has
to be what's called an "equirectangular projection".

The structure of .dsc catalogs is described in
https://en.wikibooks.org/wiki/Celestia/DSC_File

A description and examples of how to use a sphere this way are
available at
https://www.classe.cornell.edu/~seb/celestia/billboard.html#7.2

Another example of an Addon using this technique to project all-sky
maps within Celestia is available at
https://www.classe.cornell.edu/~seb/celestia/catalogs.html#3.5.9

Another way to create an all-sky grid would be to use a CMOD model
consisting only of the grid lines instead of a solid sphere.
Unfortunately, I'll have to leave that project until another day.

[Shtantan1]

Thank you so much! This seems to be just what I needed.

Added after 6 hours 15 minutes:
By modifying the wmap plugin I've been able to create a .dsc sphere with a coordinate plane. It's slightly distorted but should function fine. Image attached below.

I've been going over the documentation and messing with plug-in after plug-in all day but I just can't seem to figure out how to orient it.

I'm assuming it has something to do with the axis of the sphere. I understand it's "a unit vector specifying the orientation of the object's axis in Celestia's Ecliptic J2000 "Universal" coordinate system" but I don't know what each value represents. The wmap plug-in (which I ripped the sphere from) and the IR Milky Way plug-ins both have the same values. If they're all set to zero the sphere can't be manipulated by the "angle" parameter but if they're not all relatively similar you get some weird results?

And perhaps worth noting is that while the sphere's north/south poles start aligned to the ecliptic it's aligned at (0˚, 270˚). idk if that's to be expected. I just expected (0˚,0˚).

Code: Select all

Axis [-0.409897 0.623875 0.665406]         // values used by other plug-ins
Angle 15                           // noticeable difference... but doesn't pivot on ecliptic grid
RA 0                              // used for position; centered on Sol here
Dec 0                           // used for position; centered on Sol here
Distance 0                           // used for position; centered on Sol here
Radius 2                           // radius of sphere


So basically I've got the sphere, and now I'm trying to orient it to face perpendicular to the orbital plane of Alpha Cen.

Here's what I've got:

Code: Select all

Axis [-0.409897 0.623875 0.665406]         
Angle 0                           
RA 219.899437                        // finds Alpha Cen's RA
Dec -60.835401                        // finds Alpha Cen's declination
Distance 4.365                        // uses this parameter and the above two to place the center of the sphere around Alpha Cen
Radius 2.76e10 # 1.38e10

I imagine if one knows what the axis values control this would be moderately straightforward as the inclination and all the other good stuff is right in nearstars.stc.

[selden]

That combination of Axis and Angle values are what's known as quaternions. I have to admit I really don't grok them.

However, Celestia includes functionality so you can "eye-ball" an object's orientation, use the mouse to rotate it appropriately, and then print out the required Axis and Angle values.

See
https://www.classe.cornell.edu/~seb/celestia/billboard.html#6.0
and
https://www.classe.cornell.edu/~seb/celestia/billboard-cmod.html#6.0

[Shtantan1]

Selden, thank you so much for all your help! I really, really appreciate it. I wouldn't have been able to do this without you!

I was able to orient the sphere to a satisfactory level and am now in a position where I can take the screenshots I need.

[selden]

Great!

You're very welcome.