Batching Process in BatchLSEstimator

Hello everyone,

I am developing an Orbit Determination tool that uses the Time Difference of Arrival (TDOA) measurements. Within the OD operation, the orbit is estimated with X number of measurements. The system keeps obtaining measurements over time and it is desirable to update the estimated orbit each Y number of measurements, being Y<X. Therefore, the first estimation would be from an initial guess and X number of measurements, the second would be from the same initial guess and X+Y number of measurements, and so on. For the second and successive estimations would be ideal to use the batching process, harnessing the already performed computations of the previous estimation saving computational and time resources. However, after some trials supported on the OD tutorial and surfing through the codes, I find no clue on how the batching process is implemented (with respect to the usual Least Squares) and the availability of this feature.

Would you know if there is some available bibliography supporting the implemented batching in Orekit? If the commented feature is available, what would be the best way to implement it on my code?

Thanks for your time,

Hi @A.Garcia-Salcedo,

First, welcome to the Orekit forum.

The following reference can give you information about the batch least squares orbit determination method implemented in Orekit [1].

The feature you mentioned looks very interesting. It is like a sequential batch least squares. Unfortunately, we do not have yet that feature in Orekit … However, it can be an interesting feature to add! Can you open a feature request in our Gitlab repository?
Please note that contributions are always welcome :slightly_smiling_face:

For more information on sequential batch least squares, you can look at [2]

Kind regards,

[2]: David A. Vallado, Fundamentals of Astrodynamics and Applications, 4th Ed., Chap. 10.5 Sequential-Batch Least Squares, pp. 772-777.

Thanks Bryan,

I will make some feature request in this case.
Very helpful the answer.

Blue skies,