Dear all,
I would like to ask you some clarifications regarding how to use correctly the position angle type which can be set in a Keplerian/equinoctial orbit and in the propagator.
If I set a specific position angle type for a Keplerian or equinoctial orbit, is then necessary to set the same position angle type inside the propagator? Or can we decide to use a different position angle type inside the propagator? If this is possible, in which cases you would suggest this?
The initial orbit can be defined with a different angle type (which can also be different to the one used for storage internally by the object) than the propagation uses as independent variable. I suggest you use the default one for the latter, which is ECCENTRIC. I did some profiling a while ago and it was the most performant for elliptical trajectories.
So if I want Keplerian states as output with a TRUE anomaly, is it anyway convenient to set ECCENTRIC anomaly in the propagator, propagate in ECCENTRIC anomaly and convert it later to true anomaly?
In another case, if for example I want a Cartesian output of the propagation, is it convenient to set directly a CARTESIAN type orbit in the propagator?
Yes you can do the conversion as post process after propagation.
I would recommend using equinoctial elements over Cartesian coordinates (which are fast variables) to reduce propagation noise.