Complete OD using ITRF

[MaximeJ]

Hi @jlipinski,

Welcome to the Orekit community !

@jlipinski

I’ve infered that I need to use other, pseudo-inertial frame. Is there a way to estimate IOD using ITRF?

Currently I don’t think there is a way to use IODGibbs with a non inertial frame.
So converting your PV in ITRF to an inertial frame like GCRF is the proper way to do it.
I think it would be hard to write the IOD method so they can use non inertial frames. I’m not sure the time required to develop it would be worth it, since a simple frame conversion can do the trick.

@jlipinski

I’ve found somewhere in this forum that you can work around this error by changing orbity type to cartesian.

Yes, or maybe circular orbit. What does your orbit look like, is it a circular orbit ?

@jlipinski

Using this I’ve gotten pretty accurate orbit. RAAN, e and i are pretty spot on, a is 3 km bigger then truth but that’s okay for now.

Do you have any force model added to your propagatorBuilder. It doesn’t seem so from your code.
If so you are doing the propagation and the fitting of the orbit with a pure Keplerian model (only the point mass gravity of the Earth is taken into account).
Maybe you should add some force models if you want to have a better estimation of the orbit, like:

• A gravity model with sufficiently high degree/order
• The influence of the Sun and Moon as third bodies
• The atmospheric drag (if your satellite is low enough)
• The solar radiation pressure
  etc.

@jlipinski

why this orbit has no attributes like " getPerigeeArgument()" or “getRightAscensionOfAscendingNode()”

I think you need to cast the resulting orbit to KeplerianOrbit since the generic type Orbit doesn’t have the methods getPerigeeArgument and getRightAscensionOfAscendingNode.
See this recent answer from @bcazabonne for help on casting objects in Orekit Python.

@jlipinski

why I cant set ephmeris mode to this estimated propagator

setEphemerisMode doesn’t exist anymore in Orekit since 11.0.
The proper way to do it now is to do like in the Ephemeris Mode tutorial, that is (in Java):

// Before propagation: prepare the ephemeris generator
EphemerisGenerator generator = propagator.getEphemerisGenerator();

// Propagate
propagator.propagate(startDate, endDate)

// Get the generated ephemeris
BoundedPropagator ephemeris = generator.getGeneratedEphemeris()

@jlipinski

In summary, is there a better way of performing IOD and OD with ITRF data that I have? Maybe a way that has none of the problems mentioned above?

I think you are going in the right direction.
Doing the OD in non-inertial frame is quite more complex and prone to error (in my opinion).
What you could do is design your own measurement class that would handle the conversion from ITRF to GCRF at construction, so you wouldn’t have to explicitly do it in your main code anymore .
But it would just be moving around some part of code from a place to another, so I don’t think that it is what you’re looking for.

Hope this helps,
Maxime