Cool I take a look. Thanks. Formatting was done mostly by the RustRover IDE. Maybe I need to do a reformat at some point.
QS + Additional testing
validate_PSD
Move test helpers
nalgebra 0.34
nalgebra 0.34
Document new_weights
Hello Adrain, I seen this too. The are lots of ways the sigma transformation can be parametrised. In my UKF implementation I try follow the numerics in various publications and implementations with alpha controlling the spread. The numerics don't seem to be great and I don't see a solution at the moment. If you can build Jacobians out of your hopefully not too non-linear model the the UD filter should be numerically stable. Nalgebra 0.34 is based on Rust edition 2024 which fit nicely with my recent...