Add derivative filter and discretization

[ViktorCVS]

Overview

Adds a configurable derivative filter. It also introduces separate discretization method selectors for the I and D terms. Forward Euler remains the default to preserve existing behavior.

What was added/changed in this PR

• Added a first-order derivative filter (tf) applied to the D term.
• Introduced per-term discretization selectors: i_method and d_method (e.g., "forward_euler", "backward_euler", "trapezoidal"), with "forward_euler" as the default.
• Implemented discrete D-term updates for the supported methods, including filtered forms; falls back to unfiltered behavior when tf == 0.
• Added internal state (d_term_last_, d_error_last_, i_term_last_, aw_term_last_) to support trapezoidal (Tustin) and filtered updates.

About tests

The packages compile correctly and have passed the pre‑commit and colcon tests. New tests are being added based on formula values to compare with those from compute_command.

Related PR's

• Add derivative filter and discretization methods ros2_controllers#1909

Attachments

Full derivation of the discrete-time PID used in this PR. The document shows the step-by-step development of the equations.

ros2_control_new_discretization_feature.pdf

Final notes

I'm very open to any recommendations to improve this code.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 92.44936% with 41 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.98%. Comparing base (6a28f23) to head (2b7a909).

Files with missing lines	Patch %	Lines
control_toolbox/include/control_toolbox/pid.hpp	54.16%	19 Missing and 3 partials ⚠️
control_toolbox/src/pid_ros.cpp	68.42%	10 Missing and 2 partials ⚠️
control_toolbox/src/pid.cpp	91.66%	6 Missing and 1 partial ⚠️

Additional details and impacted files

@@               Coverage Diff               @@
##           ros2-master     #471      +/-   ##
===============================================
+ Coverage        78.90%   80.98%   +2.08%     
===============================================
  Files               29       29              
  Lines             1896     2351     +455     
  Branches           119      128       +9     
===============================================
+ Hits              1496     1904     +408     
- Misses             332      378      +46     
- Partials            68       69       +1     

Flag	Coverage Δ
unittests	80.98% <92.44%> (+2.08%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...ontrol_toolbox/include/control_toolbox/pid_ros.hpp	100.00% <ø> (ø)
control_toolbox/test/pid_ros_parameters_tests.cpp	100.00% <100.00%> (ø)
control_toolbox/test/pid_ros_publisher_tests.cpp	94.85% <100.00%> (+0.40%)	⬆️
control_toolbox/test/pid_tests.cpp	100.00% <100.00%> (ø)
control_toolbox/src/pid.cpp	81.15% <91.66%> (-2.59%)	⬇️
control_toolbox/src/pid_ros.cpp	67.65% <68.42%> (-0.22%)	⬇️
control_toolbox/include/control_toolbox/pid.hpp	57.72% <54.16%> (-3.26%)	⬇️

🚀 New features to boost your workflow:

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

[ViktorCVS]

Hi, @christophfroehlich! I still need to fix the existing tests and add new ones. However, the equations are quite large, and the compute_command function has grown a lot. Do you think it’s acceptable (with better organization), or should something be changed?

[christophfroehlich]

Hi, @christophfroehlich! I still need to fix the existing tests and add new ones. However, the equations are quite large, and the compute_command function has grown a lot. Do you think it’s acceptable (with better organization), or should something be changed?

IMHO it is acceptable to go further in this direction, thanks!

[christophfroehlich]

why have you chosen a std::string here? Wouldn't a simple enum be more lightweight here? Only thing we loose will be the human readable value in the print method without adding additional logic there. alternative would be a struct like the andiwindup strategy.

[ViktorCVS]

Thanks! I change it to struct in commit 2b7a909

[christophfroehlich]

I suppose you have not added the integration/derivative method here on purpose?

[ViktorCVS]

Yes, for now I don’t think anyone would modify the discretization methods during runtime.