Re: info on Open Robot Controller CAD : ORCCAD

Herman Bruyninckx wrote:
> I am aware of the existence of ORCCAD for quite some years already,
> and I do like it for many reasons. However, I decided not to mention
> it on the web pages explicitly, because of its license: I'm quite sure
...
> This is indeed complementary to the current Orocos scope, and
> personally I would love to see contributions from the ORCCAD
> developers. But, the license, you know...

there are indeed two points:

 - licensing issue: 
the policy at Inria is not necessarily  monolithic, indeed. In the case
of Orccad, there are historical aspects, plus the fact that Orccad is
related to various tools upon which we have no control (Ilog for the
graphical interface, VxWorks as execution platform, Esterel as reactive
kernel and verification support). So Orccad in the present state is not
likely to be able to mix up in the open source project. 
But this might not be an obstacle to usefulness, which brings me to the
following point:

 - scientific and technical complementarity:
on the one hand, the  scope in Orocos is the constitution of open source
libraries to write implementations of control laws. On the other hand,
Orccad is a tool that can construct applications by assembling
(importing) components from these libraries, and encapsulating them into
a discrete-event driven controller. Orccad then generates an executable
code, to be linked and compiled with the libraries.

So, Orccad can be used as an application builder, gluing open source
processes together. Depending on the applications targeted, if this
higher-level layer is useful, then Orccad can be of help, without
compromising the open soucre aspect of the control laws layer.


> Maybe a starting initiative for cooperation could be to brainstorm
> extensively about the design and implementation issues of ORCCAD? This
> would benefit both Orocos and ORCCAD, although we would require a
> fresh and independent implementation...

A completely new implementation would be quite ambitious ... or maybe
just parts of it?
Maybe it would be more pragmatic to consider the generated code (rather
than the code of the Orccad tool itslef), its structure and its
integrability to the Orocos framework?

Best regards,

Eric (with Daniel)

-- 
Eric Rutten                  INRIA Rhône-Alpes                projet BIP 
ZIRST, 655 avenue de l'Europe, MONTBONNOT, 38334 ST ISMIER Cedex  FRANCE 
tel:+33 (0)4 76 61 54 02                        fax:+33 (0)4 76 61 54 54 
e-mail: [email protected]       web: http://www.inrialpes.fr/bip/