[Kstars-devel] Re: kstars engine?

[Daniel-Marian Baboiu]

Applications like kstars do not "solve" the n-body problems. This part
is part is done numerically by large computing centers like JPL and made
available to the public in the form of series with thousands of terms
(do a search "JPL ephemeris").  Based on this full integration, other,
more compact series can be obtained, such as VSOP87 and Swiss Ephemeris.
A subset of the VSOP87 is presented in Jean Meeus - Astronomical
Algorithms, which is more of a cookbook than a book on celestial mechanics.

For relatively short time intervals, the n-body problem can be solved
approximately using the perturbation theory - the main term is that of
the Sun, so the orbit is elliptic, while the influence of the other
bodies is to slowly change the orbital elements. Anything more than this
is beyond the scope of the list, but you can contact me on my private
address.

Daniel Baboiu


On 10/01/2010 01:34 PM, karinqe wrote:
> Hi,
> is there any info about the algorithms the engine uses to predict the
> motions of planets? How do you "solve" the N-body problem? How precise
> is it? What do you take into account?
> I'll look into the code later for more specific information, but now
> I'm just trying to find out which algorithms are often used in
> applications like kstars and don't have enough time to search in the
> source :(
> Could you please give me some hints?
>
> (I'm doing a Bachalor's thesis on the N-body problem...well, at least
> I'm starting)
>
> Thanks in advance,
> Karin
> _______________________________________________
> Kstars-devel mailing list
> Kstars-devel at kde.org
> https://mail.kde.org/mailman/listinfo/kstars-devel
>
>