AccelerationBias CRF frame?

[Bemtgenj]

Dear developers
i have extracted from the GNSS processing (write aPrioriSolution) the accBias.X, accBias.Y and accBias.Z through the Parameterselection2IndexVector/ MatrixCalculate (reorder). Then i created a timeseries to get the epochs added to the matrix by merging the timeseries and the accBias file to get an instrument file.

I assume that this data is in the CRF frame (X,Y,Z) as it comes from the GNSS processing and no additional attitude of the satellite is provided in the last 3 aforementioned scripts / steps . So if want to get the Along-Track, Cross-Track, Radial corresponding components to follow the orbit reporting in the LOF frame i need to use the InstrumentRotate script that uses the Star Camera?

to do the rotation i need to define a type for instrument file (which type would i need to be using then out the list?)

Is my assumption correct with regards to the frame , the rotation and the type or am i missing something ? Or is the GNSS processing output in the SRF frame as the GNSS LEO processing uses the star Camera and that is what is meant in the the documentation?- This question goes along with the last point in topic #406.

One more additional question then if it is in SRF and i would like to have the data in LOF frame, and i have the SRF - CRF rotation data can i then use the instrumentRotate script to get the accBias in LOF frame

Thank you very much for your help.

Best Regards,
Jean

[AndreHauschild]

Dear Jean,

I saw your question and I was looking into a similar topic recently, so maybe I can help by providing my understanding as a fellow user of GROOPS.

I assume that this data is in the CRF frame (X,Y,Z) as it comes from the GNSS processing and no additional attitude of the satellite is provided in the last 3 aforementioned scripts / step

As far as I see, you have to specify an attitude or starCamera file for GNSS or LEO satellites in gnssProcessing, so to my understanding the accelerations are in the SRF.

So if want to get the Along-Track, Cross-Track, Radial corresponding components to follow the orbit reporting in the LOF frame i need to use the InstrumentRotate script that uses the Star Camera?

Yes, if you want the accelerations in orbital frame, you have to rotate them from SRF to orbital frame. Your star camera file gives you the rotation from SRF to CRF and then you need to generate another star camera file from CRF to orbital frame (with SimulateStarCamera), combine the two rotations (InstrumentStarCameraMultiply) and then apply the combined rotation (InstrumentRotate).

to do the rotation i need to define a type for instrument file (which type would i need to be using then out the list?)

From looking at the source of InstrumentRotate you can see it can handle ORBIT, ACCELEROMETER, GRADIOMETER, COVARIANCE3D and VECTOR3D instrument files. I guess VECTOR3D or ACCELEROMETER are suited for your case.

Or is the GNSS processing output in the SRF frame as the GNSS LEO processing uses the star Camera and that is what is meant in the the documentation?

Yes, I think this is precisely what happens! I think the documentation mentions the other case, because some program or processing steps, e.g. slrProcessing, do not require an attitude file.

One more additional question then if it is in SRF and i would like to have the data in LOF frame, and i have the SRF - CRF rotation data can i then use the instrumentRotate script to get the accBias in LOF frame

See above. You need to generate the CRF -> LOF frame rotation yourself first.

Hope that helps!

Best regards, André

[Hanbing59]

Not checked against the code yet, but it seems that the parameterizations AccBias (and also AccelerometerScaleFactors) is for satellites with accelerometer measurements, like GRACE/GRACE-FO. Not sure if that makes sense using those parameterizations for other satellites without accelerometer observations involved in the processing.

[Bemtgenj]

Dear André, Dear Hanbing,

Thank you very much for your feedback.

I agree with you that the acceleration biases are in the SRF frame as the LEO CubeSat GNSS processing uses the starCamera. Our orbit data is in LOF and i have now used the instrumentRotate script to transform from SRF to LOF with the type that i used being accelerometer. In our processing steps we have already the rotation matrices SRF-LOF, RCF LOF and SRF- CRF so that was not a problem and i have the accBias -x,y,z now in LOF.

Best regards
Jean