[JSB] added fixes to mantain the joint names order

[saikishor]

I've add some fixes to the codes to retain the proper naming order. it will maintain the same naming as the state_interfaces. Most of the time. it is same as requested.

I confirm that it is working as expected. The following are the results after testing on the TIAGo robot

With master branch:

header:
  stamp:
    sec: 27
    nanosec: 634000000
  frame_id: ''
name:
- wheel_right_joint
- torso_lift_joint
- head_2_joint
- arm_1_joint
- head_1_joint
- gripper_right_finger_joint
- arm_2_joint
- arm_3_joint
- gripper_left_finger_joint
- arm_6_joint
- wheel_left_joint
- arm_7_joint
- arm_5_joint
- arm_4_joint
position:
- -0.01883440381531276
- 0.14992120022237515
- -2.1771094615985476e-07
- 0.5000004071831876
- 2.679102557578972e-07
- 1.498072885788662e-09
- -1.3399972501658528
- -0.47999627762595054
- 1.6320817259907867e-07
- 1.3700221974953228
- 0.00558756172246877
- -5.676095425677374e-06
- -1.4900038801677784
- 1.9399743393071844
velocity:
- .nan
- -0.009820570088929573
- 0.00010381144165565768
- -0.0004520477620429137
- -0.0005884276721260586
- 0.0003904782235743167
- 0.0026640572724227402
- 0.004136899003124899
- -0.0002584158488683724
- 0.02222271007923489
- .nan
- -0.0019689225065199385
- -0.004196435823050384
- -0.025497848941894285
effort:
- .nan
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- .nan
- 0.0
- 0.0
- 0.0
---

With the fix proposed in this PR:

header:
  stamp:
    sec: 3
    nanosec: 7000000
  frame_id: ''
name:
- wheel_right_joint
- wheel_left_joint
- torso_lift_joint
- arm_1_joint
- arm_2_joint
- arm_3_joint
- arm_4_joint
- arm_5_joint
- arm_6_joint
- arm_7_joint
- gripper_left_finger_joint
- gripper_right_finger_joint
- head_1_joint
- head_2_joint
position:
- -0.02799593211424245
- 0.003465789616476478
- -1.7199045996895583e-08
- -0.0005503061596039416
- -0.0003136661241898153
- -2.6876812497356184e-05
- -5.9248231679021046e-05
- 0.000129607204049087
- 0.0012243190113494506
- 0.0002718504386338694
- 2.561374795404651e-06
- 5.5824897165386553e-08
- -5.915678524814894e-07
- -0.0003185975886408343
velocity:
- .nan
- .nan
- -1.5386795985256052e-05
- 0.028547943684709375
- -0.05535369149489766
- 0.007545245983102342
- -0.0006098150419372589
- -0.031006116599605057
- 0.005376069634232324
- -0.015109590818687155
- -0.009228197094119775
- 0.012866053358847787
- 0.0035212422027661536
- -0.008785733629921097
effort:
- .nan
- .nan
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
- 0.0
---

[codecov]

Codecov Report

Attention: Patch coverage is 81.25000% with 6 lines in your changes missing coverage. Please review.

Project coverage is 85.10%. Comparing base (d1844e6) to head (7a5b2c8).

Files with missing lines	Patch %	Lines
..._state_broadcaster/src/joint_state_broadcaster.cpp	81.25%	0 Missing and 6 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1572      +/-   ##
==========================================
+ Coverage   85.06%   85.10%   +0.03%     
==========================================
  Files         123      123              
  Lines       11712    11719       +7     
  Branches      998      996       -2     
==========================================
+ Hits         9963     9973      +10     
+ Misses       1434     1433       -1     
+ Partials      315      313       -2     

Flag	Coverage Δ
unittests	85.10% <81.25%> (+0.03%)	⬆️

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

Files with missing lines	Coverage Δ
..._state_broadcaster/src/joint_state_broadcaster.cpp	88.88% <81.25%> (-0.66%)	⬇️

... and 5 files with indirect coverage changes

🚀 New features to boost your workflow:

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

[christophfroehlich]

Shouldn't it now just use the order of the joints defined in the URDF, if we use the auto-export of the state interfaces? But with the example_1 one gets

 $ ros2 topic echo /joint_states --once
header:
  stamp:
    sec: 1741554154
    nanosec: 735820900
  frame_id: ''
name:
- joint2
- joint1
position:
- 0.0
- 0.0
velocity:
- .nan
- .nan
effort:
- .nan
- .nan

[saikishor]

Shouldn't it now just use the order of the joints defined in the URDF, if we use the auto-export of the state interfaces? But with the example_1 one gets

If the joints parameter is not defined, then it publishes in the original order of the state interfaces

[christophfroehlich]

Thanks for the follow-ups and the docs, just minor suggestions there

[mhubii]

Thank you for the effort and also considering URDF order!

I have tested this on mock hardware for lbr_fri_ros2_stack inside an Ubuntu 24.04 Docker container using option 1), 2), and 3).

ros2_control master branch source build.

• Option 1: ✅
• Option 2: ✅
• Option 3: ✅

However, this error occurred frequently:

[ros2_control_node-2] [ERROR] [1743008902.700347616] [lbr.controller_manager]: Caught exception of type : St18bad_variant_access while updating controller 'joint_state_broadcaster': std::get: wrong index for variant

Not sure this is related.

[saikishor]

However, this error occurred frequently:

[ros2_control_node-2] [ERROR] [1743008902.700347616] [lbr.controller_manager]: Caught exception of type : St18bad_variant_access while updating controller 'joint_state_broadcaster': std::get: wrong index for variant

Not sure this is related.

I'm not sure if this has something to do with the joint_state_broadcaster. IMO for now, we can discard this