I have ephemeris data for state vectors of a satellite in a J2000.0 coordinate system. This data is already filtered and cleaned.

The data sample rate ranges between 1 second and 60 seconds, and I want to resample the data down to <10 milliseconds with a constant, regular time step.

I have already managed to interpolate using methods from scipy.interpolate with cubic interpolation; interp1d and griddata.

This leads me to the question of how to choose between various methods to resample/interpolate ephemeris state vectors. Certainly the interpolation of tabular ephemerides is something that astronomers and orbital dynamicists need to do and so there must exist some guidelines or best practices or recognized-better or -worse ways to so this.

Question: How to choose between methods to resample/interpolate ephemeris state vectors? Are there methods recognized as superior for orbital data?

  • $\begingroup$ I'm voting to close this question as off-topic because this is a software coding question and unrelated to the source of the data $\endgroup$ – Carl Witthoft May 2 '19 at 14:21
  • $\begingroup$ This is not related to software coding at all, I'm look for a method or models to improve the sampling, independent from the implementation. I gave scipy as an example. $\endgroup$ – Matty May 2 '19 at 14:29
  • $\begingroup$ There are a huge variety of ways to interpolate data, so I consider this too broad. $\endgroup$ – StephenG May 2 '19 at 15:09
  • $\begingroup$ This is a great question, and it is on-topic both here and in Space Exploration SE as well. The close-voters don't recognize how common it is for astronomers to interpolate ephemerides, or that you've already done the interpolation a few ways. Rather than help you refine the wording of the question, they chose to close it (I wouldn't have chosen that route). I've adjusted the wording to make your question more specific and amenable to the Stack Exchange format. $\endgroup$ – uhoh May 4 '19 at 0:40
  • $\begingroup$ I don't know if there are more people who can answer here or in Space SE, so I am not sure if it's better to vote to re-open here, or wait to see if a moderator decides to migrate there. In the mean time, I'm curious what happens if you compare the results for the two methods how close they agree with each other. Also if you can describe how you are using the resampled data and how you would decide which results were better, that will be a big help. Do you just need least error on position, or best smoothness (lowest excursions in higher order derivatives, e.g. acceleration and above)? $\endgroup$ – uhoh May 4 '19 at 0:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.