[ros-users] urdf and tf questions

Tully Foote tfoote at willowgarage.com
Wed Sep 29 18:29:11 UTC 2010


Ben,
The common way to use tf is to publish
link -> link -> link
and use the joint angles to compute the transform between links.  It is
easier if you make the origin of each link at the joint.

As Lorenz said, if you are using a URDF, the robot_state_publisher will
subscribe to joint_states and publish the tf frames for the links in the
URDF.

If you want to just view the URDF you could load it into gazebo.

Tully

On Wed, Sep 29, 2010 at 8:33 AM, David Lu!! <davidvlu at gmail.com> wrote:

> If you don't want to worry about publishing joint angles yet, and just
> want to visualize your model, check out easy_tf_publisher in my
> urdf_tools stack.
> http://www.ros.org/wiki/easy_state_publisher
> Or if you want to manually set the joint angles, try joint_controller
> and robot_state_publisher.
>
> -David!!
>
> 2010/9/29 Lorenz Mösenlechner <moesenle at in.tum.de>:
> > Hi,
> >
> > the common way is to run a robot_state_publisher which takes the joint
> > states on a topic and publishes the transforms. You find the
> corresponding
> > tutorial here:
> >
> http://www.ros.org/wiki/robot_state_publisher/Tutorials/Using%20the%20robot%20state%20publisher%20on%20your%20own%20robot
> >
> > Lorenz
> >
> >> What is the typical usage when publishing transforms to match a urdf
> >> file?   For example, the urdf file has the standard form:
> >>
> >> Link1 -> Joint1 -> Link2 -> Joint2 -> Link3
> >>
> >> In my code, I tf broadcast the joint angles.  (i.e. Link1 -> Joint1 and
> >> Link2 -> Joint2).  But must I also broadcast the link transforms even
> >> though they never change?  (i.e. Joint1 -> Link2 and Joint2 -> Link3).
> >> Is there some way to avoid this?
> >>
> >> But what I would REALLY like to do for now is just view the URDF file in
> >> rviz.  But despite all origins being specified in the file, rviz says it
> >> can't find the transforms between the objects.  Is there a simple URDF
> >> file viewer (in boxturtle)?
> >>
> >> Thanks,
> >> -Ben
> >>
> >
> > --
> > Lorenz Mösenlechner            | moesenle at in.tum.de
> > Technische Universität München | Boltzmannstr. 3
> > 85748 Garching bei München     | Germany
> > http://ias.cs.tum.edu/         | Tel: +49 (89) 289-26910
> >
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1.4.9 (GNU/Linux)
> >
> > iEYEARECAAYFAkyjWeAACgkQruZZyu9ksInbKwCeJuZ/FNuePMpd2pLfUpgcPW7K
> > AzQAn1EHFoyY+rIrLPwLRSWqVZE53Xro
> > =lKpk
> > -----END PGP SIGNATURE-----
> >
> > _______________________________________________
> > ros-users mailing list
> > ros-users at code.ros.org
> > https://code.ros.org/mailman/listinfo/ros-users
> >
> >
> _______________________________________________
> ros-users mailing list
> ros-users at code.ros.org
> https://code.ros.org/mailman/listinfo/ros-users
>



-- 
Tully Foote
Systems Engineer
Willow Garage, Inc.
tfoote at willowgarage.com
(650) 475-2827
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ros.org/pipermail/ros-users/attachments/20100929/32ff6c82/attachment-0003.html>


More information about the ros-users mailing list