added: covestim in MovingHorizonEstimator now supports SteadyKalmanFilter
#248
+105
−49
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
franckgaga
changed the title
SteadyKalmanFilter support for covestim in SteadyKalmanFilter support for covestim in MovingHorizonEstimator
franckgaga
changed the title
SteadyKalmanFilter support for covestim in MovingHorizonEstimatorcovestim in MovingHorizonEstimator now supports SteadyKalmanFilter
Benchmark Results (Julia v1)Time benchmarks
Memory benchmarks
|
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #248 +/- ##
==========================================
+ Coverage 98.60% 98.63% +0.02%
==========================================
Files 26 26
Lines 4443 4458 +15
==========================================
+ Hits 4381 4397 +16
+ Misses 62 61 -1 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
I now use `pkgversion`
franckgaga
commented
Sep 11, 2025
I wouldn't bother with the additional complexity, for the simple reason that people updating packages in order to get access to a new feature, such as that added in this PR, tend to be on the latest version of most packages and therefore won't need the fix. Explicit version management like this only tends to be warranted when there is a significant user base that lags in their adoption of new package versions for some reason, and these users critically need bugfixes.
Yes, I noticed that the are solver in MatrixEquations returns the covariance matrix for the direct form but the gain matrix for the indirect form. The function You can read the changes to the docstring here |
To avoid the intricate `pkgversion` branching.
|
Alright, I will revert the |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This new feature allows two things:
covestimis aSteadyKalmanFilter. Note that the defaultcovestimfor the MHE with aLinModelis still a time-varyingKalmanFilter.NonLinModel. The user just needs to construct aSteadyKalmanFilterwith a dummyLinModelwith the same number of estimated statenx̂than the MHE, and fix the arrival covariance usingsetstate!as above.The
covestimargument is also now displayed when pretty-printing aMovingHorizonEstimator, in the "arrival covariance" field:Note that it relies on the newextra=Val(true)arguments inControlSystemsBase.jl, introduced in v1.18.2. This PR hence bumps the compat entry toControlSystemsBase = "1.18.2"edit: Come to think of it, I don't this that's a good idea to bump the compat of
ControlSystemsBase.jljust for this niche feature. I now rely onpkgversion(ControlSystemsBase)to pass or not theextrakeyword argument tokalman. Do you see any drawback to this workaround @baggepinnen ?edit 2: Also I notice that the computed steady-state Kalman gain with
direct=trueis not the same before and after v1.18.2, you corrected a mistake in the formula @baggepinnen ?