RE: [Orekit Developers] Feature #34 (Getting the acceleration computed by a ForceModel)

[Tanguy Yannick <Yannick.Tanguy@cnes.fr>]

-----Message d'origine-----
De : MAISONOBE Luc [mailto:luc.maisonobe@c-s.fr] 
Envoyé : mardi 24 janvier 2012 20:03
À : orekit-developers@orekit.org
Objet : Re: [Orekit Developers] Feature #34 (Getting the acceleration computed by a ForceModel)

>>
>> This email follows the discussion about feature #34.
>>
[..]

>> - for the other force models (dissipative force models : sun radiation 
>> pressure and atmospheric drag), the problem is different because we 
>> need to fix some vehicule characteristics (shape, mass).
>> So the modification we made is only used for validation purpose and 
>> should stay in a derived class.

>I don't understand fully this. Could you explain a little further ?

In fact we developed specific method for our validation tests, in which a simple vehicle is "hard-coded". This method had to be public because of our unit tests, but can't be integrated in OREKIT (it won't be used by other users). Maybe the best is to keep it in a derived class, used only in specific unit tests. This choice can be discussed, but it's a compromise :  trying to validate non conservative forces is quite tough. If we fix all parameters (surface, mass, coefficient), we only validate the drag formula. But if we stay at the highest level, results depend on the vehicle modelisation…

Yannick

>Nevertheless, I think this feature would be a real improvement to Orekit.

>Luc