control_performance_analysis crashes with segfault

kasper-helm commented 1 year ago

Checklist

[X] I've read the contribution guidelines.
[X] I've searched other issues and no duplicate issues were found.
[X] I'm convinced that this is not my fault but a bug.

Description

I have seen control_performance_analysis crash when the idx_prev_wp_ pointer is not initialized, and is dereferenced on https://github.com/autowarefoundation/autoware.universe/blob/main/control/control_performance_analysis/src/control_performance_analysis_core.cpp#L378

this happens when the ego is either beyond acceptable_max_distance_to_waypoint_ from the trajectory, or the heading error is larger than acceptable_max_yaw_difference_rad_, which causes closest_segment to be empty, and the idx_prev_wp_ to remain uninitialized

Expected behavior

The node should safely handle the case where closest_segment is empty

Actual behavior

[ERROR] [control_performance_analysis_exe-24]: process has died [pid 216, exit code -11, cmd ...

Steps to reproduce

I am using a pre-defined fixed Trajectory message to test the Autoware control component without the planning component
This issue occurs when the localization initializes while the vehicle is somewhat far (> a few meters) from the trajectory

Versions

No response

Possible causes

No response

Additional context

No response

stale[bot] commented 1 year ago

This pull request has been automatically marked as stale because it has not had recent activity.

kyoichi-sugahara commented 7 months ago

@kasper-helm Thank you so much for your report and sorry for late response :sob: If it's possible, could you give me the information how to reproduce problem? (e.g. which version of autoware you are using, how did you launch autoware, which comman did you run to publish trajectory.

kasper-helm commented 7 months ago

I think I fixed it by adding the following line above https://github.com/autowarefoundation/autoware.universe/blob/main/control/control_performance_analysis/src/control_performance_analysis_core.cpp#L378

if (!idx_prev_wp_) { return std::nullopt; }

kyoichi-sugahara commented 7 months ago

@kasper-helm Wonderful thanks! Sorry for not supporting well :cry: If you create PR, I will review the PR :+1:

stale[bot] commented 5 months ago

This pull request has been automatically marked as stale because it has not had recent activity.

autowarefoundation / autoware.universe