Robotics software: the future should be open

Author: Herman Bruyninckx, Department of Mechanical Engineering, K.U.Leuven, Belgium

This position paper introduces a number of Problem Claims about the pitiful state of practice in software for robotics (and for all kinds of engineering domains in general). The paper also presents Solution Claims, whose realization can lead to a long-term, macro-economically optimal solution, both for industry and academia.

Pretty good assessment of the situation, thanks Herman. Some specific comments:

• Robotics Software Journal: Maybe add “focus on real-world integration” to the editorial policy
• SysML Profiles for Robotics: Let's start collecting some links here, so that we can assess what these things imply.
  • http://www.sysml.org/
  • http://www.omgmarte.org/
  • http://www.papyrusuml.org/
• Apply FOS Business Model: There are already a lot of robotics-related projects on http://sourceforge.net/ and similar sites, but is there any robotics-specific community platform? Maybe RAS would be a good place to host such a thing.

Some comments on the comment above:

* Maybe the RAS site could host an index to these projects rather than the projects themselves, like the old robotics FAQ used to do, and provide a taxonomy and maybe even ratings. This could be a community, wiki-based, process. Of course RAS could also host projects but given the ubiquity of sourceforge and Google code that may not be sensible.

* As for SysML and MARTE, maybe a tutorial article in RAM to explain what they are, not in an abstract sense, but how to use them for a real (probably simple) robotics problem. What tools exist to move from model representation to executable code in your favourite language working on your favourite robotics API?