fix(voxel_based_compare_map): temporary fix pointcloud transform lookup

[badai-nguyen]

Description

• This temporary override input_indices_callback to change lookup tf from latest to pointcloud's header stamp.
• To fix this issue: Compare_map_segmentation degradation at high speed or during turns #10290
• This could remove when lookup_time could be select when calling transform_pointcloud from autoware_utils or new ManagedTransformBuffer

Related links

• TIER IV internal link

Parent Issue:

• Link: Compare_map_segmentation degradation at high speed or during turns #10290

How was this PR tested?

• By logging_simulator on internal rosbag

Before	After (a large number pc were filtered)
Screencast from 2025年03月19日 08時33分35秒_before.webm	Screencast from 2025年03月19日 08時33分35秒_before.webm

• By internal evaluator (WIP):

Notes for reviewers

None.

Interface changes

None.

Effects on system behavior

None.

[github-actions]

Thank you for contributing to the Autoware project!

🚧 If your pull request is in progress, switch it to draft mode.

Please ensure:

• You've checked our contribution guidelines.
• Your PR follows our pull request guidelines.
• All required CI checks pass before marking the PR ready for review.

[codecov]

Codecov Report

Attention: Patch coverage is 30.43478% with 16 lines in your changes missing coverage. Please review.

Project coverage is 25.91%. Comparing base (660ae1a) to head (915a4fa).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...tation/src/voxel_based_compare_map_filter/node.cpp	30.43%	12 Missing and 4 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #10299      +/-   ##
==========================================
- Coverage   25.92%   25.91%   -0.01%     
==========================================
  Files        1382     1382              
  Lines      106811   106860      +49     
  Branches    40907    40928      +21     
==========================================
+ Hits        27690    27698       +8     
- Misses      76418    76456      +38     
- Partials     2703     2706       +3     

Flag	Coverage Δ		*Carryforward flag
differential	21.83% <30.43%> (?)
differential-cuda	18.86% <30.43%> (?)		Carriedforward from 8a17324
total	25.92% <ø> (-0.01%)	⬇️	Carriedforward from 8a17324

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[mojomex]

@badai-nguyen There seems to be some previously non-existent misalignment in some regions now. It looks like those parts of the pointcloud have an angular offset to the map.

Before	After

I wonder if it is an issue with this Pr, with other bugs or delays in the inputs, or if this is a visualization issue (delay between pointclouds).

[mojomex]

Thank you for the PR. I have a few timing-related comments 🙇

[YoshiRi]

LGTM and it worked within my v4.1.0 environment

[YoshiRi]

Pointcloud number decreased and it looks better now. But there seems to be some part is not removed yet.

Before	After

[badai-nguyen]

There seems to be some previously non-existent misalignment in some regions now. It looks like those parts of the pointcloud have an angular offset to the map.

@mojomex
Thanks for your insight review.
I think transform back output of compare_map_filter from map -> baselink still use the default convert with latest tf
Which might cause the misalignment. I will fix it.

[badai-nguyen]

@badai-nguyen There seems to be some previously non-existent misalignment in some regions now. It looks like those parts of the pointcloud have an angular offset to the map.

Before After

I wonder if it is an issue with this Pr, with other bugs or delays in the inputs, or if this is a visualization issue (delay between pointclouds).

@mojomex
I fixed this misalignement at 15546f6
The result after update as here:

video:
Screencast from 2025年03月20日 00時17分53秒_after_fix_back_tf.webm

[yukkysaito]

This seems like a very temporary workaround. I think it would be better to make it unnecessary to comment out the above before merging. What do you think?

[yukkysaito]

Or do you mean that the commented-out feature is intended to be deprecated?

[badai-nguyen]

@yukkysaito I'm sorry, it was my typo. I fixed 89523c9
By the way, we also started a discussion about deprecation of unused filters. I will open the disucssion later.