Repository navigation
Preserve predicted EKF priors and support missing measurements - #348
Open
xixihaha-ha wants to merge 1 commit into
Open
xixihaha-ha wants to merge 1 commit into
xixihaha-ha wants to merge 1 commit into
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
ExtendedKalmanFilter.predict_updatesaves the oldself.x/self.Pas its prior, rather than the local predicted state/covariance. It also evaluates the observation functions forz=Noneand then fails, although the method documents a prediction-only step for missing measurements.Save the computed predicted values. For a missing measurement, retain them as current/posterior state and return before calling observation functions, matching separate
predictandupdate(None)calls.Tests compare combined and separate steps with control input for present/missing observations, including all prior/posterior snapshots. A separate test rejects any observation-function call for a missing measurement.
Validation: all 3 new tests fail on unchanged main and pass after the fix. Including existing EKF tests: 5 passed (Python 3.11 / NumPy 1.26).
AI assistance was used to investigate, implement and test this change.