Quantcast

Re: [ros-users] rosrecord compilation error

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [ros-users] rosrecord compilation error

tfoote
Jack,
Sorry for the delay, but I've posted a fuller answer onto the tf FAQ page http://www.ros.org/wiki/tf/FAQ#Why_don.27t_frame_ids_use_ROS_namespaces.3F

Tully

On Fri, Dec 11, 2009 at 10:39 AM, Josh Faust <[hidden email]> wrote:
I think the answer is that there is already a hierarchy defined by the tf graph in the child/parent relationships, vs. ROS names where the hierarchy is defined by the name.  Realistically you could certainly use the tf prefix to "push down", just like you can with namespaces in ROS, since it's all just a string.

Tully can respond in more detail if I'm wrong.

Josh



I appreciate the pointer, Josh.

I see how adding that parameter on a per-robot basis allows multiple
robots to share the same TF name space.  But, it seems rather
surprising (to me) that names for objects in a completely hierarchical
TF graph are themselves only two-level.

Perhaps the reasons have so far escaped me.  Why are frame IDs not
full-fledged ROS names?  All the benefits of "pushing down" node and
topic names have value for frames of reference, too, don't they?
--
 joq


------------------------------------------------------------------------------
Return on Information:
Google Enterprise Search pays you back
Get the facts.
http://p.sf.net/sfu/google-dev2dev

_______________________________________________
ros-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ros-users




--
Tully Foote
Systems Engineer
Willow Garage, Inc.
[hidden email]
(650) 475-2827

------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
_______________________________________________
ros-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ros-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: [ros-users] rosrecord compilation error

Jack O'Quin
On Wed, Jan 20, 2010 at 2:03 PM, Tully Foote <[hidden email]> wrote:
> Jack,
> Sorry for the delay, but I've posted a fuller answer onto the tf FAQ page
> http://www.ros.org/wiki/tf/FAQ#Why_don.27t_frame_ids_use_ROS_namespaces.3F

That helps me to grasp at least some of the issues.  Not sure I
completely understand, but it is clear that tf names need to connect
different parts of the robot, where node namespaces often correspond
to individual components or collections of components.  So, just
pushing the names down (as I naively assumed) would not work.

Using a two-level namespace readily allows names with either per-robot
scope or global scope.  Most of the examples that come to mind fit
that model.  For example, two robots may have different odometry
frames (robot1/odom and robot2/odom), each related to the global /map
frame by separate localization transforms.

Intuitively, I expected there to be other tf name scopes of interest,
perhaps for movable arms, like /robot1/left/elbow, and
/robot2/right/wrist.  I suppose the hierarchical parameter search for
tf_prefix allows that, using a recursive definition with each
subordinate name prefixed by the values of any higher-level tf_prefix
values.  Setting those parameters seems tricky and somewhat confusing,
however.

Maybe I am imagining more complexity than the problem requires.

Regards,
--
 joq

------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
_______________________________________________
ros-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ros-users
Loading...