Noticed that attitude.cpp:satYawGalIov() always returns false for E101 (E11), E102 (E12) and E103(E19) in a particular dataset which seems odd. These are in the GALILEO-1 block which appears intended to have modelled attitude implemented. Is it expected that these all have an invalid modelled attitude over a period of a few hours.
Compared results using the CODE precise attitude product which does include an attitude for these satellites and noted a significant difference in results vs using the Ginan attitude (Nominal fallback). With these satellites excluded the differences are trivial. Might this point to an issue with the Ginan modelled attitude.
Noted that these can be excluded when the model is not valid by not including NOMINAL as a fallback in the config attitude:sources.
Btw also noted issues with GLONASS even after excluding R807 (R19) for which the model attitude does not appear to be implement in Ginan. Perhaps there is some difference in the models compared to CODE, or perhaps CODE just solved significantly different attitudes compared to the expected models in post processing. Noted some minor transient result differences for BDS, and no significant differences for GPS and QZSS.
Noticed that attitude.cpp:satYawGalIov() always returns false for E101 (E11), E102 (E12) and E103(E19) in a particular dataset which seems odd. These are in the GALILEO-1 block which appears intended to have modelled attitude implemented. Is it expected that these all have an invalid modelled attitude over a period of a few hours.
Compared results using the CODE precise attitude product which does include an attitude for these satellites and noted a significant difference in results vs using the Ginan attitude (Nominal fallback). With these satellites excluded the differences are trivial. Might this point to an issue with the Ginan modelled attitude.
Noted that these can be excluded when the model is not valid by not including NOMINAL as a fallback in the config attitude:sources.
Btw also noted issues with GLONASS even after excluding R807 (R19) for which the model attitude does not appear to be implement in Ginan. Perhaps there is some difference in the models compared to CODE, or perhaps CODE just solved significantly different attitudes compared to the expected models in post processing. Noted some minor transient result differences for BDS, and no significant differences for GPS and QZSS.