Re: Octave and optimal control

[John W. Eaton]

On 15-Aug-2007, James Sherman Jr. wrote:

| On 8/14/07, Paul Smith <address@hidden> wrote:
| >
| > On 8/12/07, Paul Smith <address@hidden> wrote:
| > > Dear All,
| > >
| > > I would like to know whether Octave can do optimal control, i.e., like
| > > RIOTS does in Matlab:
| > >
| > > http://www.schwartz-home.com/~adam/RIOTS/
| > >
| > > Thanks in advance,
| > >
| > > Paul
|
| I'm just gonna take a stab at this and ask, have you tried using RIOTS with
| octave?  Since octave is designed to be matlab compatible, unless there is
| some functions RIOTS uses that aren't implemented in octave or can't be
| found here http://octave.sourceforge.net/packages.html, then I'd imagine it
| would work.
| 
| I really don't know much about this though, so please ignore if this is
| completely off the mark.

I looked at the web page and all I see there are some self-extracting
.exe files.  I don't have a convenient way to run those, so I haven't
looked at the files, but are all the sources available for these?  If
not, and if RIOTS depends on some MEX files that are only available in
binary form, then I don't think there is much hope for running them in
Octave.  OTOH, if there is source available, you'd probably have a
better chance.  Also, what is the license for this software?  It's
rather inconvenient that one must download and run the .exe to extract
the license info...  In any case, from looking at the manual, it
appears to use NPSOL, which is not free software, so it seems doubtful
that RIOTS could be free software.  That's too bad.

jwe