Velocity in reference frame transformations

[juan...@gmail.com]

Hello all,

I'm Juan Luis Cano, a long time follower of the Astropy project using it for some personal projects. I would like to ask you about a somewhat complex issue so I thought the developer mailing list would be the best place - please feel free to redirect me elsewhere if it's more convenient.

I am the sole developer of poliastro, an interplanetary astrodynamics Python library relying heavily in many astropy features. On the other hand, I joined efforts with other developers to create a core astrodynamics package, and we will surely make use of astropy too. One of our main concerns involves the transformation between reference frames, which is already implemented in astropy.coordinates but lacks an essential feature for us: velocity transformations.

I understand that this feature could be out of scope of the astropy project, but still wonder what would be the best way to implement it ourselves. I was thinking of subclassing the main reference frames, adding some information like transformation derivatives or velocity of rotation and overloading the `transform_to` method, but haven't sketched a detailed design yet. If the astropy developers could provide some feedback, ideas or guidance I would greatly appreciate it.

Thanks in advance and congratulations for such an amazing source of daily inspiration :)

Regards,

Juan Luis

[Marten van Kerkwijk]

Hi Juan,

we *really* would like to have velocity transformations as part of astropy.coordinates, so this is very much *in* scope.  Indeed, it is part of our current planning. Any help will be very much appreciated!
I should add that at present we're thinking very much of distant objects, i.e., of proper motions and radial velocities -- see https://github.com/astropy/astropy/issues/4344 -- but we would like things to work inside the solar system as well

​

All the best,

Marten

[Adrian Price-Whelan]

Hi Juan --

I should also add that a group of us interested in the future of
astropy.coordinates are planning to have a telecon on Tuesday 9 Feb at
13:00 Eastern time. If you're interested in attending, please let Erik
Tollerud know (cc'd).

Thanks,
ADrian

> --
> You received this message because you are subscribed to the Google Groups
> "astropy-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to astropy-dev...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.



--
Adrian M. Price-Whelan ~ Columbia University ~ http://adrian.pw

[Juan Luis Cano]

Hi Adrian and Erik,

Thanks a lot for the invitation. If I understood correctly it's via
Google Hangouts at 18:00 UTC, correct? I would love to attend if possible.

Kind regards,

Juan Luis Cano