Topic/extend whole body state ros publisher#16
Conversation
cmastalli
left a comment
There was a problem hiding this comment.
The main requested change is to support ROS 2 as well. This requires several changes, and I have written many of them. There are other minor changes.
Please let me know when you're done! Also check that the CI passes.
|
@cmastalli IMO PR ready to merge |
cmastalli
left a comment
There was a problem hiding this comment.
The main change requested is to follow the Pinocchio convention when interfacing with inertial parameters. This needs to be done in both publisher and subscriber.
Additionally, there are minor changes needed for ROS2 and handling default arguments appropriately.
Finally, the PR misses update/get inertia parameters in the whole-body trajectory publisher and subscriber.
Please handle this ASAP. It is a good idea to finish earlier what was started.
|
@Sergim96 Could you investigate why ROS 2 jobs are not passing? It is worth mentioning that Pinocchio fails to import in ROS 1 (see #14). This is known although I don't remember the reason now. In short, after passing ROS 2 jobs, we could deactivate the inertial test. A better alternative would be to adapt the cmake to important inertial and whole-body tests for ROS 2 jobs only. It would be very useful if you could handle this in this PR. Please ping me when this is done. Let's try to finish this asap. Thanks! |
|
HI @cmastalli now the ROS2 humble is passing, but rolling is failing to setup the environment in this line: and as you said ROS1 fails because of pinocchio, I can disable it but the test for the whole body state also uses pinocchio, so I expect it to fail too |
33fc9c0 to
8842a02
Compare
…nertial Parameters
…tial parameters add pub/sub
8842a02 to
e4c5046
Compare
Hi @cmastalli
This PR brings a new message type to be able to communicate dynamic parameters between different nodes.
It also implements the corresponding publisher/subscriber and unittests.
Additionally I have extended the WholeBodyStateRosPublisher in order to publish accelerations as well #5