Re: [Orca-robotics-devel] Re: [Playexrstage-developers] Common robotics data exchange format...?

On Mon, 16 Aug 2004, Alex Brooks wrote:

> Could you explain a bit more what your vision is?  Perhaps with some
> example data structures?

2D position and orientation, 3D position and orientation, joint values
of robot arms and humanoids, representation of kinematic chains (simple,
but a little bit more complex for humanoids), 2D and 3D velocities and
accelerations. That's basically it.

> I'm confused because you talk about defining data structures, which to
> me indicates making some choices about representation, eg. XML, packed
> binary, IDL, etc, but this doesn't seem to be your intent.
No, it isn't: as soon as we succeed in specifying how we represent
things the same way, the implementation of the representations (XML,
IDL, etc.) will be easy to do; and wrappers (if an application would
need them) are easy too.

> Secondly, could you explain why you say the data representation is more
> important than the communications infrastructure?  Surely both have to
> match for two systems to inter-operate.

Yes. But in practice, there are lots of working communication
infrastructures already, but no common data representations :-)

Herman

-- 
   K.U.Leuven, Mechanical Engineering, Robotics Research Group
<http://people.mech.kuleuven.ac.be/~bruyninc> Tel: +32 16 322480