RE: Modelica

> The Modelica people decided to steer away from the Bond graph ideas
> behind modelling of dynamical systems, which is in my opinion a poor
> decision, because the ideas behind bond graphs (with or without the
> possibly clumsy notation :-) ) really represent the physics of
> mechanical systems...
> Another `deficiency' (which could easily be solved by Modelica) is the
> data structure: an XML-based language would have been much more
> versatile.
> 
> But mind you: at the time being, I still haven't found anything better
> than Modelica, and I am not squarely against it. Moreover, there are
> already nice examples of modelled robots in Modelica (thanks to Martin
> Otter), including models of drives and controlers.

Well, the complete bondgraph geometrical setting is implemented in 20sim.
It has its own language and the underlying structure is well geometrically
defined using screws. 
 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Dr.ir. Stefano Stramigioli
Associate Professor

Faculty of Electrical Engineering
Drebbel Institute

P.O. Box 217
NL-7500 AE Enschede
The Netherlands

Tel. +31 (53) 4892794/4892606 
Fax. +31 (53) 4892223 

Email [email protected]
WWW: http://www.rt.el.utwente.nl/smi

ICQ: 7864978
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%