Satellite Mathematical Model

[10k10]

Hi Eric, how are you? I'm here again.

I've been working on attitude estimation using QUEST to estimate the quaternions and EKF to filter it and estimate my non-measured states (angular velocities). I can track quaternions very well even when there are a lot of noise. However, I can't track the angular velocities, even without any noise. As you know, as better my model is, better are the estimation of the velocities, so I compared my discrete model with 42, using the same initial conditions and moments of inertia and verified a difference in \omega_y showed in the pic below:

I'm using something like that to my satellite model:

Now I'm wondering what is wrong with my model, that's why I can't track the velocities well (my guess).

Any guess about it? Is 42's satellite model similar (intrinsically)?

Best regards,

Robinson.

[10k10]

It was my mistake converting data from the output file to Matlab.