fix(ndt_scan_matcher): fix map_update_module structure

[SakodaShintaro]

Description

In this pull request, the structure of map_update_module in ndt_scan_matcher has been changed to optimize processing.

Context

(1) Currently, map_update_module continuously outputs ERROR logs due to its inability to subscribe to ekf odometry until ndt_scan_matcher estimates the initial pose. Outputting ERROR in this normal sequence is not ideal. To avoid this, a structural change was required because map_update_module needs to check the state of ndt_scan_matcher.

(2) Upon reviewing the code, I identified another issue: map_update_module subscribes to ekf odometry independently of ndt_scan_matcher. This redundancy is inefficient. Given that the ekf odometry topic is published at 50Hz, subscribing to it twice could negatively impact CPU usage.

(3) Also, because map_update_module creates callback groups in its constructor, it is difficult to understand how many callback groups there were in the entire ndt_scan_matcher.

Changes

• Moved ekf subscriber and timer callbacks to ndt_scan_matcher_core.cpp.
• Changed the map_update_module to no longer manage the current location and focus only on updating the map.
• Changed the name from main_callback_group to sensor_callback_group.
• Some minor refactoring, such as removing unused arguments.

Tests performed

The logging_simulator has been verified to run with the same accuracy as before on AWSIM data with GT. It also performs well when dynamic_map_loading is used for divided maps.

Effects on system behavior

There are no large effects on system behavior. There is a possibility that the CPU load has improved slightly because EKF's double subscription is eliminated.

Pre-review checklist for the PR author

The PR author must check the checkboxes below when creating the PR.

• I've confirmed the contribution guidelines.
• The PR follows the pull request guidelines.

In-review checklist for the PR reviewers

The PR reviewers must check the checkboxes below before approval.

• The PR follows the pull request guidelines.

Post-review checklist for the PR author

The PR author must check the checkboxes below before merging.

• There are no open discussions or they are tracked via tickets.

After all checkboxes are checked, anyone who has write access can merge the PR.

[KYabuuchi]

Thanks for your contribution. Please check my following comments. 🙏

[codecov]

Codecov Report

Attention: 28 lines in your changes are missing coverage. Please review.

Comparison is base (acf7e51) 15.28% compared to head (fa3c0b5) 15.32%.
Report is 10 commits behind head on main.

Files	Patch %	Lines
...ion/ndt_scan_matcher/src/ndt_scan_matcher_core.cpp	0.00%	12 Missing and 10 partials ⚠️
...ization/ndt_scan_matcher/src/map_update_module.cpp	0.00%	5 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #5915      +/-   ##
==========================================
+ Coverage   15.28%   15.32%   +0.03%     
==========================================
  Files        1749     1747       -2     
  Lines      120356   119816     -540     
  Branches    36708    36305     -403     
==========================================
- Hits        18401    18356      -45     
+ Misses      81305    81092     -213     
+ Partials    20650    20368     -282     

Flag	Coverage Δ		*Carryforward flag
differential	3.07% <0.00%> (?)
total	15.32% <ø> (+0.03%)	⬆️	Carriedforward from 6a3a3a5

*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.

[SakodaShintaro]

Also, the is_activated_ variable is accessed from multiple threads as shown below.

• Read/write by sensor_callback_group thread
• Read in initial_pose_callback_group and timer_callback_group_ threads

I made it atomic. Please let me know if there is any misunderstanding.

(When I made it an atomic variable, I got a warning from clang-tidy if I didn't initialize it in the member initializer lists, so I changed the initialization location.)

[KYabuuchi]

LGTM
I think the two additional exclusivity controls make sense. 👍 (for is_activated_ and latest_ekf_position_ )