RE: Open source robot controllers

> From: Herman Bruyninckx [mailto:[email protected]]
> Subject: RE: [Orocos] Open source robot controllers
> 
> 
> On Mon, 24 Sep 2001, Tiller, Michael (M.M.) wrote:
> 
> [...]
> > >
> > > It could be interesting for you to take a look at 
> modelica.org, for
> > > the modelling language. I do have some objections against 
> the precise
> > > way they are doing things, but it's by far the most 
> advanced thing at
> > > this instant.
> > >
> > A minor clarification.  Modelica covers the kinds of requirements
> > you describe for connectors.  In addition, there are a (currently
> > informal, I think) set of annotations that allow Modelica models and
> > hierarchies to be represented and connected graphically.  These are
> > implemented by Dymola (http://www.dynasim.se) and MathModelica
> > (http://www.mathcore.com).
> 
> Thanks for the clarification. But they don't have any open source
> implementation available, as far as I know?

Well, this really depends on what you mean.  The entire Modelica effort is open source.  The standard library, the specification and the meeting minutes are all made available for free.

But, you need to appreciate the difference between Modelica and tools that implement Modelica.  There are no open source simulation tools for Modelica (at least ones that I am aware of).  Modelica is just a language, not a tool.

So when you say "they" it is important to clarify who "they" are.  If "they" is the Modelica Association, then everything is open source.  If "they" means the vendors, then there is not an open source simulator/code generator.

Of course, somebody could implement an open source simulator.  It is just that nobody has done this (yet).

> > Finally, Modelica goes beyond just requiring pairs of effort and
> > flow variables.  For example, in multi-body systems it is often
> > necessary to express these quantities as vectors.
> 
> This is of course a necessity for robotics.
> 
> > Furthermore, in
> > thermodynamic system (my area) we often have far more complex
> > connectors where the flow and non-flow variables are not balanced.
> 
> What do you mean with `not balanced'?

Well, for example...We are concerned with the flow of a particular gas.  The state of the gas is described by the pressure, temperature and chemical composition (mass fractions).  Typically, this would be 6 scalars.  Any flow of the gas through a branch in the system is represented by two quantities: mass flow rate and heat flow rate (i.e. convected energy flow rate).  Finally, there are certain quantities (e.g. specific enthalpy) which are propagated in a "dataflow" way (kind of hard to describe why).

In other words, connectors are used to indicate a more general concept of information sharing between components.  Sometimes the information is "dataflow" (using the 'input' and 'output' qualifiers in Modelica), sometimes it is conserved (using the 'flow' qualifier) and in some (admittedly rare cases) it is a hybrid of all of the above.

These hybrid schemes should really be avoided unless they are absolutely necessary.  However, when they are necessary it is very nice that they are allowed.

> Herman

--
Mike