perf(autoware_ndt_scan_matcher): remove evecs_, evals_ of Leaf for memory efficiency #9281
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
taisa1
requested review from
anhnv3991,
KYabuuchi,
Motsu-san,
RyuYamamoto,
SakodaShintaro,
TaikiYamada4 and
YamatoAndo
as code owners
github-actions
bot
added
the
component:localization
|
Thank you for contributing to the Autoware project! 🚧 If your pull request is in progress, switch it to draft mode. Please ensure:
|
… threshold (autowarefoundation#9279) fix computation of maximum lane changing length threshold Signed-off-by: mohammad alqudah <[email protected]> Signed-off-by: taisa1 <[email protected]>
Signed-off-by: taisa1 <[email protected]>
taisa1
force-pushed
the
perf_ndt_scan_matcher_shrink_leaf
branch
from
d766f86
to
158928f
Compare
taisa1
requested review from
mkquda,
rej55,
kosuke55,
maxime-clem,
satoshi-ota,
shmpwk,
TomohitoAndo,
tkimura4 and
zulfaqar-azmi-t4
as code owners
github-actions
bot
added
the
component:planning
Signed-off-by: taisa1 <[email protected]>
planning/behavior_path_planner/autoware_behavior_path_lane_change_module/src/scene.cpp
Outdated
Show resolved
Hide resolved
…/autoware.universe into perf_ndt_scan_matcher_shrink_leaf
github-actions
bot
removed
the
component:planning
SakodaShintaro
removed request for
satoshi-ota,
tkimura4,
maxime-clem,
zulfaqar-azmi-t4 and
mkquda
SakodaShintaro
added
the
tag:run-build-and-test-differential
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #9281 +/- ##
==========================================
- Coverage 29.23% 29.22% -0.01%
==========================================
Files 1335 1335
Lines 102959 102963 +4
Branches 39954 39958 +4
==========================================
- Hits 30102 30093 -9
- Misses 69976 69995 +19
+ Partials 2881 2875 -6
*This pull request uses carry forward flags. Click here to find out more. ☔ View full report in Codecov by Sentry. |
SakodaShintaro
approved these changes
Nov 11, 2024
YamatoAndo
approved these changes
Nov 11, 2024
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.
Description
By investigating the memory usage during scan matching, I found that the memory consumption was dominated by
std::vector<MultiVoxelGridCovariance<pcl::PointXYZ>::Leaf>in addition to the map itself.The following figure shows the memory usage when executing autoware's
ndt_scan_matcherwithsample-map-rosbag. The bottom area corresponds tostd::vector<MultiVoxelGridCovariance<pcl::PointXYZ>::Leaf>(7.9MB).The struct
Leafcontains threeEigen::Matrix3d, andsizeof(Leaf)is 288, so having this for each voxel may consumes a lot of memory with large maps.In this PR, I propose to remove
evecs_(Eigen::Matrix3d) andevals_(Eigen::Vector3d) fromLeaf. These variables are only used to calculatecov_and not be used directly. This will reducesizeof(Leaf)from 288 to 192.When tested with
sample-map-rosbag, the memory usage ofstd::vector<MultiVoxelGridCovariance<pcl::PointXYZ>::Leaf>was reduced from 7.9MB to 5.2MB.Related links
How was this PR tested?
I did a regression test on https://github.com/tier4/ndt_omp and got the following result:
Notes for reviewers
None.
Interface changes
None.
Effects on system behavior
None.