feat: goal planner checking border #10725

Owen-Liuyuxuan · 2025-05-28T04:57:37Z

Description

Filter out objects that are separated from the ego vehicle by road borders, preventing unnecessary path planning considerations for objects on the opposite side of road infrastructure.

Algorithm

(New)

(before 6/1)

Before

After

How was this PR tested?

Notes for reviewers

None.

Interface changes

None.

Effects on system behavior

None.

github-actions · 2025-05-28T04:57:54Z

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 · 2025-05-28T05:14:46Z

Codecov Report

Attention: Patch coverage is 0% with 53 lines in your changes missing coverage. Please review.

Project coverage is 16.58%. Comparing base (7ff3aa9) to head (4ff8162).
Report is 16 commits behind head on main.

Files with missing lines	Patch %	Lines
...are_behavior_path_goal_planner_module/src/util.cpp	0.00%	50 Missing ⚠️
...th_goal_planner_module/src/goal_planner_module.cpp	0.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #10725      +/-   ##
==========================================
+ Coverage   15.94%   16.58%   +0.63%     
==========================================
  Files        1343     1351       +8     
  Lines       99736   100502     +766     
  Branches    32880    33188     +308     
==========================================
+ Hits        15907    16672     +765     
- Misses      71635    72166     +531     
+ Partials    12194    11664     -530

Flag	Coverage Δ		*Carryforward flag
daily	`17.33% <ø> (+<0.01%)`	⬆️	Carriedforward from f4353f2
daily-cuda	`15.95% <ø> (+<0.01%)`	⬆️	Carriedforward from f4353f2
differential	`8.28% <0.00%> (?)`
total-cuda	`15.95% <ø> (+<0.01%)`	⬆️	Carriedforward from f4353f2

*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.

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp

kosuke55 · 2025-05-29T11:06:08Z

Overall, it looks good! Thanks a lot!! I just left a few minor comments.

kosuke55 · 2025-05-29T11:19:12Z

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp

+    // Check if a road border separates ego from object
+    if (!crosses_road_border(ego_point, obj_point, road_border_segments)) {


It seems okay to ignore for now, but this is an edge case of the current algorithm.

Actually there are more complex cases, so I tried to implement a more strict method for filtering objects.

For now, it will loop through the footprint of objects like the following graph:

For bouding box object, I compute the footprint using the pose + rotation + dimensions. For polygon object, I directly use the footprint in the message. For the cylinder object, I still use the centerpoint.

thanks!!
I can not open the link

@kosuke55

Sorry I copied the private user link

great image!! I'm grad if you added to README.md too

I added the graph in the README with the fix commit. Please check it out.

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp

kosuke55 · 2025-06-04T02:47:52Z

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp

+    // Get object center position
+    const auto & obj_pose = object.kinematics.initial_pose_with_covariance.pose;
+    const Point2d obj_center_point{obj_pose.position.x, obj_pose.position.y};
+


Suggested change

// Get object center position

const auto & obj_pose = object.kinematics.initial_pose_with_covariance.pose;

const Point2d obj_center_point{obj_pose.position.x, obj_pose.position.y};

not used now?

I think so, since we are obtaining polygon directly.

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/README.md

kosuke55

LGTM! thanks a lot!

…utowarefoundation#10692) Signed-off-by: Makoto Kurihara <[email protected]> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Mamoru Sobue <[email protected]> Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

Owen-Liuyuxuan · 2025-06-04T09:34:11Z

sorry the DCO suggestion creates shit.

kosuke55 · 2025-06-04T09:36:04Z

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/README.md

+Before these margins are applied, objects are filtered based on their location relative to road boundaries (linestring with [roader_boarder](https://github.com/fzi-forschungszentrum-informatik/Lanelet2/blob/master/lanelet2_core/doc/LinestringTagging.md#as-lane-boundary) type ).
+  Objects located on the opposite side of relevant road borders from the ego vehicle are excluded from the collision check process. This ensures that the planner focuses only on objects that pose a potential risk within the intended driving corridor.


Suggested change

Before these margins are applied, objects are filtered based on their location relative to road boundaries (linestring with [roader_boarder](https://github.com/fzi-forschungszentrum-informatik/Lanelet2/blob/master/lanelet2_core/doc/LinestringTagging.md#as-lane-boundary) type ).

Objects located on the opposite side of relevant road borders from the ego vehicle are excluded from the collision check process. This ensures that the planner focuses only on objects that pose a potential risk within the intended driving corridor.

Before these margins are applied, objects are filtered based on their location relative to road boundaries (linestring with [road_border](https://github.com/fzi-forschungszentrum-informatik/Lanelet2/blob/50239be48aa5a911d7347fdb828ee79dde2b2666/lanelet2_core/doc/LinestringTagging.md#as-lane-boundary) type ).

Objects located on the opposite side of relevant road borders from the ego vehicle are excluded from the collision check process. This ensures that the planner focuses only on objects that pose a potential risk within the intended driving corridor.

sorry for my typo, and fixed hash.

Sorry the DCO suggestion use force push and make the PR a chaos.. I am not sure what to do now.

@kosuke55 This is fixed with the new PR..
#10755

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

* merging updates in #10725 Signed-off-by: YuxuanLiuTier4Desktop <[email protected]> * fix hash and typo, precommit Signed-off-by: YuxuanLiuTier4Desktop <[email protected]> --------- Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

…autowarefoundation#10755) * merging updates in autowarefoundation#10725 Signed-off-by: YuxuanLiuTier4Desktop <[email protected]> * fix hash and typo, precommit Signed-off-by: YuxuanLiuTier4Desktop <[email protected]> --------- Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

Owen-Liuyuxuan requested review from danielsanchezaran and kosuke55 as code owners May 28, 2025 04:57

github-project-automation bot added this to Software Working Group May 28, 2025

Owen-Liuyuxuan requested review from kyoichi-sugahara, soblin, satoshi-ota, shmpwk, TomohitoAndo and tkimura4 as code owners May 28, 2025 04:57

github-project-automation bot moved this to To Triage in Software Working Group May 28, 2025

github-actions bot added the component:planning Route planning, decision-making, and navigation. (auto-assigned) label May 28, 2025

Owen-Liuyuxuan added run:build-and-test-differential Mark to enable build-and-test-differential workflow. (used-by-ci) tag:require-cuda-build-and-test labels May 28, 2025

kosuke55 reviewed May 29, 2025

View reviewed changes

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp Outdated Show resolved Hide resolved

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp Outdated Show resolved Hide resolved

kosuke55 reviewed May 29, 2025

View reviewed changes

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/src/util.cpp Outdated Show resolved Hide resolved

kosuke55 reviewed May 29, 2025

View reviewed changes

kosuke55 reviewed Jun 2, 2025

View reviewed changes

kosuke55 reviewed Jun 4, 2025

View reviewed changes

github-actions bot added the type:documentation Creating or refining documentation. (auto-assigned) label Jun 4, 2025

kosuke55 reviewed Jun 4, 2025

View reviewed changes

planning/behavior_path_planner/autoware_behavior_path_goal_planner_module/README.md Outdated Show resolved Hide resolved

kosuke55 approved these changes Jun 4, 2025

View reviewed changes

mkuri and others added 6 commits June 4, 2025 18:33

feat: implement to filter objects on the other sides of road border

eaeb9b2

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

use parameter

53e4590

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

improve goal planner mechanism

87ee721

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

formatting

dd292f5

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

precommit

9916b65

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

Owen-Liuyuxuan requested review from MasatoSaeki, technolojin, YoshiRi, ktro2828, badai-nguyen, ogrex, yukkysaito, Shin-kyoto, ito-san, TetsuKawa, mkuri, taikitanaka3, xtk8532704, anhnv3991, domrobotec, Motsu-san, SakodaShintaro, TaikiYamada4 and YamatoAndo as code owners June 4, 2025 09:33

kosuke55 reviewed Jun 4, 2025

View reviewed changes

Owen-Liuyuxuan closed this Jun 4, 2025

github-project-automation bot moved this from To Triage to Done in Software Working Group Jun 4, 2025

Owen-Liuyuxuan added a commit to Owen-Liuyuxuan/autoware.universe that referenced this pull request Jun 4, 2025

merging updates in autowarefoundation#10725

86321b6

Signed-off-by: YuxuanLiuTier4Desktop <[email protected]>

		// Check if a road border separates ego from object
		if (!crosses_road_border(ego_point, obj_point, road_border_segments)) {

	// Get object center position
	const auto & obj_pose = object.kinematics.initial_pose_with_covariance.pose;
	const Point2d obj_center_point{obj_pose.position.x, obj_pose.position.y};

		Before these margins are applied, objects are filtered based on their location relative to road boundaries (linestring with [roader_boarder](https://github.com/fzi-forschungszentrum-informatik/Lanelet2/blob/master/lanelet2_core/doc/LinestringTagging.md#as-lane-boundary) type ).
		Objects located on the opposite side of relevant road borders from the ego vehicle are excluded from the collision check process. This ensures that the planner focuses only on objects that pose a potential risk within the intended driving corridor.

feat: goal planner checking border #10725

feat: goal planner checking border #10725

Uh oh!

Conversation

Owen-Liuyuxuan commented May 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Algorithm

Before

After

Related links

How was this PR tested?

Notes for reviewers

Interface changes

Effects on system behavior

Uh oh!

github-actions bot commented May 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov bot commented May 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kosuke55 commented May 29, 2025

Uh oh!

kosuke55 May 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

kosuke55 left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Owen-Liuyuxuan commented Jun 4, 2025

Uh oh!

kosuke55 Jun 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Owen-Liuyuxuan commented May 28, 2025 •

edited

Loading

github-actions bot commented May 28, 2025 •

edited

Loading

codecov bot commented May 28, 2025 •

edited

Loading

kosuke55 May 29, 2025 •

edited

Loading

kosuke55 left a comment •

edited

Loading

kosuke55 Jun 4, 2025 •

edited

Loading