Make the ManifoldUpdate much more efficient #134
Merged
Conversation
This file contains 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
Codecov Report
@@ Coverage Diff @@
## main #134 +/- ##
==========================================
- Coverage 91.50% 90.89% -0.62%
==========================================
Files 24 24
Lines 942 955 +13
==========================================
+ Hits 862 868 +6
- Misses 80 87 +7
Continue to review full report at Codecov.
|
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Outdated suggestions cannot be applied.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.
Previously, each IEKF iteration allocated a bunch of new matrices, which hurt the performance: A solve with a ManifoldUpdate callback took ~15ms for a system that is solved in ~1.5ms without callback.
Now the number of fresh allocations is greatly reduced and the solve with callback takes only ~2.7ms.
This could still be improved, e.g. by using some of the already existing cache matrices or by requiring that the residual function is in-place, or even that it has to have a specific output shape.
EDIT: The issue with having the same dimensionality with the residual function as the ODE itself is that the measurement covariance becomes singular. Right now I'm not sure how to best solve this.