I am trying to simulate measurements with modifiers beyond just noise/offset. When generating measurement data using the MeasurementBuilder classes I can see how to introduce errors using the noiseSource parameters. There is also the method to to add EstimationModifiers that will be passed on to each created measurement. The modifiers to seem to be just pass through values however and don’t seem to affect the “observed” measurements.
Is there a recommended approach for how to incorporate an EstimationModifier, say Bias, such that it is accounted for in the observed measurement?
Orekit is a gem btw, just getting wrapped around the axle on a few parts!
Welcome to the Orekit forum!
Estimation modifiers are use to take into consideration some effects on measurement modelling. These effects can be for instance atmospheric effects (i.e. tropospheric and ionospheric effects) or biases (i.e. station and satellite biases). These modifiers can be used to take into consideration the previous perturbations in measurement modelling during an orbit determination or to generate noisy measurements.
For measurement generation, a tutorial is available in the Orekit tutorial project. I recommend you to have a look on this tutorial. The
getBuilder() method of the tutorial performs the initialisation of a Range measurement builder with the addition of the
EstimationModifier (I think it is what you are looking for).
It uses YAML to initialise the mandatory data for measurement generation (i.e. orbit propagator used for measurement generation, measurement modifier setting, etc.). Do not hesitate to download the Orekit tutorial project and play with that tutorial (To run the tutorial, you must run the
PerformanceTesting class). Please note that an orbit determination is performed using the previous generated measurements.
If you have other questions, do not hesitate to ask them.