-
Notifications
You must be signed in to change notification settings - Fork 42
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Add MuJoCo by FZI and AlgoryX #250
Conversation
@Mergifyio backport humble iron |
✅ Backports have been created
|
Thanks for spotting and referencing that implementation! However, I'm afraid that's a little robot dependent at the moment. As far as I'm concerned, MuJoCo needs a specific configuration .xml that specifies the robot to simulate. We could adapt/change it to provide a more generic interface between MuJoCo and ros2_control so that users can simulate arbitrary robots (manipulators). I didn't continue this idea lately, but I'm interested in supporting/implementing if that's of use for the community. |
For the scope of this PR it is fine like it is. It should just give a reference implementation for users on control.ros.org I have a personal interest in bringing MuJoCo and ros2_control closer due to a research project I'm working on. But I haven't tried it out yet and can't comment on the limitations. @traversaro pointed me to google-deepmind/mujoco#1432, so I think there is more work to be done to provide a seamless integration like with Gazebo (classic). Once I have the time to work on that, I'd be happy to discuss future steps to find a more general solution. |
Just to complement the link, I think most of people using URDF with MuJoCo either manually or automatically change the URDF before the loading in MuJoCo, or the resulting |
de72d48
to
a78341d
Compare
a78341d
to
569ae73
Compare
(cherry picked from commit 55de5ea)
(cherry picked from commit 55de5ea)
ros2_control system interface for MuJoCo. Not tested but looks great!