[ros-users] Draft REP: Coordinate Frames for Humanoids Robots

Tully Foote tfoote at willowgarage.com
Thu Jan 12 19:25:40 UTC 2012


I too think this REP is a good idea.  I have some feedback below:

For brevity I suggest changing the names to use l_ and r_ for left and
right prefixes instead of left_ and right_.  It saves a lot of typing and
makes debugging strings fit in smaller displays better.  It is less
explicit, but I have not seen the ambiguity become an issue in existing
models.

I also note that the frame higherarchy does not have some frames such as
elbows and knees.  I suggest that the specification be adjusted to allow
for intermediate frames between the higherarchy.  And also make sure that
this REP can be extended for more complicated robots.  Also other
consideration I would suggest is to consider trying to follow as much of
existing conventions which are not codified yet, because there exists a lot
of code already using the conventions.  For example I would suggest
switching, chest -> torso, and hand -> gripper.  There's some level of
improvement which can be made to existing conventions, however I would
suggest the default be existing conventions and changes be justified.  I
know I'm using the PR2 specific conventions, however I also think that
there is the most code using those conventions already.

Tully

On Thu, Jan 12, 2012 at 1:01 AM, Matthieu Herrb <matthieu.herrb at laas.fr>wrote:

> On Wed, Jan 11, 2012 at 12:06:54PM +0100, Thomas Moulard wrote:
> > Hi all,
> > Attached is a draft defining naming conventions and semantic meaning
> > for coordinate frames of legged robots used within ROS.
>
> A couple of small nits...
>
> >
> > left_hand and right_hand
> > ''''''''''''''''''''''''
> >
> > The coordinate frame called ``left_hand`` defines the position and
> > orientation of the left arm end effector.
> >
> > If the end effector is a grasping device, the frame should be located
> > at the recommended object grasping location. The frame orientation is
> > defined as X the axis which should be collinear with the object axis
> > during the grasping process, Y the axis going "toward" the reached
> > object and Z the axis orthogonal to the (X,Y) plan.
> >
> >
> > The coordinate frame called ``right_wrist`` defines the equivalent
>                                      ^^^^^
>                                      pasto
> > frame for the right arm.
> >
> >
> > Frame Authorities
> > -----------------
> >
> > Unlike mobile bases, the transformation from ``base_link`` to
> > ``base_footprint`` is not rigid on humanoid robots and is broadcasted by
> > an external node.
>
> Rigidity is also not always true for mobile platforms. Especially
> outdoor chassis with suspended wheels can have home complex
> geometrical transformation here, which can't always be approximated as
> a constant...
>
>
> --
> Matthieu Herrb
> _______________________________________________
> 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/20120112/a999f926/attachment-0004.html>


More information about the ros-users mailing list