Re: [Rugged Users] Direct location- frames

[MAISONOBE Luc <luc.maisonobe@c-s.fr>]

vfonsny@gmail.com a écrit :

> Hello,

Hi,

> I'm trying to use the direct location but I have some trouble with  
> the frames.
> In the tutorial, it is said that the line of sight of the sensor has to be
> given in local frame such as X is the velocity, Z points towards the  
> Earth and
> Y is fixed to have a right handed coordinate system. I suppose that the
> positions and velocities used are those in the inertial frame.
> Thus I thought that with a rotation equal to the one given by the transform
> from the inertial frame to the non inertial one and raws directions with  the
> values:
> (0, [a range of near 0 values ], 1) the spacecraft would always sees
> the Earth. It is not the case and I can not find how I have to define the
> rotation. Can you tell me what is wrong ?

Hi,

Rugged expect the LOS directions always in satellite frame. In the tutorial,
for the sake of simpler explanation, we have assumed this satellite frame was
aligned with a local orbital frame defined in terms of satellite position
and velocity. This does not need to be always true, and in real cases,  
attitude
is often offset from this simple case.

So in real cases, there are two related things that must be set:

 - the LOS directions in spacecraft frame (often completely fixed)
 - spacecraft frame orientation with respect to inertial frame,
   this is given by attitude, specified with the setTrajectory
   method in the RuggedBuilder

If the attitude is not consistent with the LOS directions, you may not
see the Earth.

In the tutorial, the attitude is specified directly from a set of quaternions,
which were precomputed elsewhere. These quaternions are valid only at the
tutorial date and for the tutorial position and velocity, they correspond to
a consistent data set.

In real operational systems, quaternions are often provided as image metadata,
typically from a star tracker telemetry.

For simulation and test purposes, quaternions can also be computed internally,
thanks to the underlying Orekit library: you can set up an  
AttitudeProvider that
will compute the theoretical quaternion given the spacecraft position  
and velocity.

So my guess is that LOS/attitude/position are not consistent if you don't see
the Earth.

best regards,
Luc

> Thank you.