use gz-physics#283 to implement joint_states/effort feedback #186
Conversation
|
I tried it and it works properly, couldn't find any problems |
christophfroehlich
force-pushed
the
fix-effort-state
branch
from
c83f289
to
70eb141
Compare
|
We are currently integrating Gazebo and would like to have the efforts from simulation. How is this PR going? |
|
I'll take a quick look this weekend whether I can get this PR cleaned up and submitted. |
|
This should do it (further testing strongly recommended): andreasBihlmaier@59ef8f0 I'll try to either update this PR or create a new one in the next days. |
andreasBihlmaier
force-pushed
the
fix-effort-state
branch
from
17fd89c
to
59ef8f0
Compare
|
I did a bit of quick testing wrsp to joint axis. The effort seems to be reported as expected. |
|
I tested this. It works for us. Thanks a lot @andreasBihlmaier |
|
@ahcorde this looks good to me. Any reason for not merging? |
christophfroehlich
force-pushed
the
fix-effort-state
branch
from
15830d5
to
1b1c12f
Compare
|
https://github.com/Mergifyio backport rolling humble iron |
✅ Backports have been created
|
(cherry picked from commit cc66e73)
(cherry picked from commit cc66e73) # Conflicts: # gz_ros2_control/src/gz_system.cpp
(cherry picked from commit cc66e73) # Conflicts: # ign_ros2_control/src/ign_system.cpp
…430) (cherry picked from commit cc66e73) Co-authored-by: Andreas Bihlmaier <[email protected]>
Fixes
joint_states/effortbeing zero for joints with<state_interface name="effort" />in combination withjoint_state_broadcaster/JointStateBroadcasterby using theJointTransmittedWrenchcomponent introduced by gazebosim/gz-physics#283.The joint axis is hard coded to
z. I did not yet look into reading out the actual<axis>of the joint and extracting the respective force/torque component of the wrench.