Euclidean Debug#3731
Open
sunghyuneun wants to merge 18 commits into
Open
Conversation
sunghyuneun
requested review from
Andrewyx,
sauravbanna and
williamckha
as code owners
Formatting accidentally deleted a requirements_lock This reverts commit 7f7fb0f.
Andrewyx
reviewed
May 22, 2026
Contributor
Andrewyx
left a comment
Andrewyx
left a comment
There was a problem hiding this comment.
Nice work! Left some comments
Comment on lines
+27
to
+30
| double fr_x = 0.03485, fr_y = -0.06632; | ||
| double fl_x = 0.03485, fl_y = 0.06632; | ||
| double bl_x = -0.04985, bl_y = 0.05592; | ||
| double br_x = -0.04985, br_y = -0.05592; |
Contributor
There was a problem hiding this comment.
- We should store these constants in
robot_constants.halongside and also update the diagram for the wheel orientations in the meantime
| euclidean_to_wheel_velocity_D_.transpose(); | ||
| } | ||
|
|
||
| #else |
Contributor
There was a problem hiding this comment.
- Either delete the legacy code or add a TODO & issue for removing this old code in the future.
| std::cos(b_br), std::sin(b_br), (br_x * std::sin(b_br) - br_y * std::cos(b_br)); | ||
| // clang-format on | ||
|
|
||
| // Calculate Pseudo-inverse dynamically |
Contributor
There was a problem hiding this comment.
We know all of these values statically, is there some way to compute everything before runtime in this case? Can this entire constructor be constexpr?
Comment on lines
+173
to
+180
| EXPECT_NEAR(std::abs(calculated_wheel_speeds[FRONT_RIGHT_WHEEL_SPACE_INDEX]), | ||
| expected_lever_arm, 0.001); | ||
| EXPECT_NEAR(std::abs(calculated_wheel_speeds[FRONT_LEFT_WHEEL_SPACE_INDEX]), | ||
| expected_lever_arm, 0.001); | ||
| EXPECT_NEAR(std::abs(calculated_wheel_speeds[BACK_LEFT_WHEEL_SPACE_INDEX]), | ||
| expected_lever_arm, 0.001); | ||
| EXPECT_NEAR(std::abs(calculated_wheel_speeds[BACK_RIGHT_WHEEL_SPACE_INDEX]), | ||
| expected_lever_arm, 0.001); |
Contributor
There was a problem hiding this comment.
|
|
||
| // Because the wheels are offset, their speed at -1 rad/s equals their | ||
| // exact physical lever arm (in meters) multiplied by the negative velocity. | ||
| double expected_lever_arm = 0.0749; |
Comment on lines
+16
to
+17
| auto p = robot_constants_.front_wheel_angle_deg * M_PI / 180.; | ||
| auto t = robot_constants_.back_wheel_angle_deg * M_PI / 180.; |
Contributor
There was a problem hiding this comment.
No need for auto here, and also statically known values
Andrewyx
reviewed
May 23, 2026
Contributor
Andrewyx
left a comment
Andrewyx
left a comment
There was a problem hiding this comment.
Almost there, left some comments!
Comment on lines
+44
to
+55
| // X position of centre of front wheel | ||
| float front_wheel_x_mag; | ||
|
|
||
| // Y position of centre of front wheel | ||
| float front_wheel_y_mag; | ||
|
|
||
| // X position of centre of rear wheel | ||
| float back_wheel_x_mag; | ||
|
|
||
| // Y position of centre of rear wheel | ||
| float back_wheel_y_mag; | ||
|
|
Contributor
There was a problem hiding this comment.
what is mag? Note that we should ideally keep units in the name too e.g. front_wheel_x_meters -> // X position of centre of front wheel relative to the center of the robot.
| .wheel_radius_meters = 0.03f}; | ||
| .wheel_radius_meters = 0.03f, | ||
|
|
||
| .expected_lever_arm = 0.0749f}; |
Contributor
There was a problem hiding this comment.
How is this computed? Also can you label this somewhere in the ASCII diagram or something to better explain what this value points to?
Contributor
StarrryNight
left a comment
StarrryNight
left a comment
There was a problem hiding this comment.
nice work. Left some comment on code quality and inversing matrices.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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
Rewrote the inverse kinematics logic for increased readability and correctness.
Revised wrong tests to be correct.
Testing Done
Also edited
euclidean_to_wheel_test.cpp, and passed all tests in it.Resolved Issues
Fixes #3722
Length Justification and Key Files to Review
Review Checklist
It is the reviewers responsibility to also make sure every item here has been covered
.hfile) should have a javadoc style comment at the start of them. For examples, see the functions defined inthunderbots/software/geom. Similarly, all classes should have an associated Javadoc comment explaining the purpose of the class.TODO(or similar) statements should either be completed or associated with a github issue