Iterated Posterior Linearization FIlter (IPLF) #1016
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
mrcfon
requested review from
akenyon and
sdhiscocks
and removed request for
a team
sdhiscocks
reviewed
May 16, 2024
Member
sdhiscocks
left a comment
sdhiscocks
left a comment
There was a problem hiding this comment.
Overall looks good.
Doc strings need improving in places, minor formatting; but main thing is some suggestions (not tested) to avoid creating lots of new updater instances.
a) inherit from LinearGaussian, rather than being its sibling class b) be true generalisation of LinearGaussian, i.e., generating the measurement matrix either based on the 'mapping' specification or using the explicitly specified measurement matrix
a) added the docstring and updated the comments b) included a user-specified function that may be useful to deal with issues due to angle wrapping c) replaced np.linalg.inv with np.linalg.pinv d) minor style corrections, e.g., around @
…ugmentedGaussianMeasurementPrediction) to preserve the info about the model used to produce a measurement prediction
a) included explicit specification of the slr_func, should an alternative definition be required b) homogenised the iterations such that the first iteration is not performed separately as an UKF update c) switched from the 'while' loop to the 'for' loop for improved clarity d) removed re-definition of the measurement model in hypothesis.measurement, which then cancelled the need to save/recover the hypothesis, as no info is manipulated -- except for adding/updating a measurement prediction e) included re-definition of the measurement prediction through a custom class (AugmentedGaussianMeasurementPrediction), such that the linearised model and its parameters could be preserved (an outlook for smoothing)
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1016 +/- ##
==========================================
- Coverage 93.61% 93.50% -0.12%
==========================================
Files 203 203
Lines 13051 13141 +90
Branches 2666 2684 +18
==========================================
+ Hits 12218 12287 +69
- Misses 588 596 +8
- Partials 245 258 +13
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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 PR adds the following classes:
The Iterated Posterior Linearization FIlter (IPLF) described in [1] is obtained by iteratively linearising the measurement function using statistical linear regression (SLR) with respect to the posterior (rather than the prior), to take into account the information provided by the measurement.
A running example can be found here.
[1] Á. F. García-Fernández, L. Svensson, M. R. Morelande and S. Särkkä, "Posterior Linearization Filter: Principles and Implementation Using Sigma Points," in IEEE Transactions on Signal Processing, vol. 63, no. 20, pp. 5561-5573, Oct.15, 2015, doi: 10.1109/TSP.2015.2454485.