Closed ChristopherRabotin closed 4 months ago
Visit the preview URL for this PR (updated for commit ef77dcc):
https://nyx-rustdoc--pr340-331-measurement-upda-xc7kd37s.web.app
(expires Sat, 27 Jul 2024 03:12:02 GMT)
🔥 via Firebase Hosting GitHub Action 🌎
Sign: d8e2a55934352d850c15d11866c39eb2d2e029be
Summary
Until this PR, the spacecraft state generator made the completely incorrect assumption that each component of a state was iid. Obviously, this is wrong. This PR aims to fix this by removing the Generator structure and replacing it with the Multivariate Normal.
This also removes the ability to sample from anything but a normal distribution: in my experience, this is all that's ever done.
Architectural Changes
Remove
Generator
structure.New Features
No change
Improvements
No change
Bug Fixes
No change
Testing and validation
Sampling from a full Kalman Filter estimate and from a fully defined Cartesian state space is checked to meet the 99.7% criteria with 1000 samples. There is following validation work in #339 . However the "robust" OD tests have switched to this new dispersion method very successfully.
Documentation
This PR does not primarily deal with documentation changes.