Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Restore virtual joint type to fixed in SRDF #86

Merged
merged 2 commits into from
Jun 25, 2021

Conversation

captain-yoshi
Copy link

The virtual joint type set to floating seems to break things when doing benchmarks with the moveit_ros_benchmarks package.

The first issue I had was with STOMP with cartesian goal constraints which you can read here. Changing the joint type to fixed resolves the problem but it may not be the culprit.

The second issue is only occurring when I populate a scene with meshes (tried different meshes and reduced to only one). The left image is a joint constraint goal and the blue line is the path retrieved from STOMP. The left end of the path is the start state and the right end of the segment is the goal state. The left image is with the fixed virtual joint and is the correct planning scene. The right image was taken with the virtual joint as floating. This changes the TF from the world frame to the panda_link0 frame for an unknown reason. This also affects OMPL planners. Using this scene with SolidPrimitives and the virtual joint set to float dosen't produce the error, only meshes.

The comment in the SRDF says "considered fixed with respect to the robot" but the joint is floating... The first commits had the fixed type but was changed to floating here.

I did not investigate further but I can replicate the behavior everytime.

@v4hn
Copy link
Contributor

v4hn commented Jun 23, 2021

The comment in the SRDF says "considered fixed with respect to the robot" but the joint is floating... The first commits had the fixed type but was changed to floating here.

You are right, this can be seen as a regression due to my update of the config package.
Please also remove the corresponding tf publisher in the demo mode.

However, the error itself has nothing to do with it and this basically shows us that we need tests with a planar/floating joint config!
Changed behavior when different object types are present is an obvious error that should be addressed.

Can you please provide a minimal example for the error and document it in a new issue in the moveit repository?
If possible only use the core moveit repository + the panda config with the virtual joint set to floating.

@captain-yoshi
Copy link
Author

Please also remove the corresponding tf publisher in the demo mode.

Done.

However, the error itself has nothing to do with it and this basically shows us that we need tests with a planar/floating joint config!

I agree! It doesn't occur with Solid Primitives so it shouldn't happen with meshes.

Can you please provide a minimal example for the error and document it in a new issue in the moveit repository?

I was afraid you were going to say that! Will do after our meeting.

@v4hn v4hn merged commit 02e1ac3 into moveit:master Jun 25, 2021
@v4hn
Copy link
Contributor

v4hn commented Jun 25, 2021

Thank you for the cleanup 👍

@captain-yoshi
Copy link
Author

FYI the fixed joint removes these error messages when launching the demo:

[ERROR] [1624678879.084098588]: Multi-DOF-Joints not supported. The robot won't move. virtual_joint
[ERROR] [1624678879.084120838]: Multi-DOF-Joints not supported. The robot won't move.

@v4hn
Copy link
Contributor

v4hn commented Jun 26, 2021 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants