Kalman filter code?

[Matthew R Wette]

Robin Becker <robin at jessikat.fsnet.co.uk> writes:

> In article <8mbna2$fav$1 at news.rz.uni-karlsruhe.de>, Bjoern Giesler
> <un4e at rzstud1.rz.uni-karlsruhe.de> writes
> >Hi,
> >
> >does anyone have a Python implementation of a Kalman filter? Otherwise, I'm
> >going to write one... but I'd rather spare myself the work (and reading of
> >papers) if there's already something out there.
> >
> >Regards,
> >               --Björn
> I did a lot of stuff on Kalman filters many years ago in Fortran. I
> suppose you are talking either constant-system discrete/continuous time
> filter discrete time time varying is harder and I never did the full
> blown continuous time varying case.
> 
> You will certainly be needing NumPy.

> -- 
> Robin Becker

Check the following reference.  Conventional Kalman filtering
algorithms are numerically unstable.  The algorithms in this 
paper come out to <=5 lines of MATLAB code.

       M.Verhaegen, P. Van Dooren, "Numerical Aspects of Different Kalman
       Filter Implementations," IEEE TAC AC-31(1986)10, p. 907-917.

Matt

-- 
Matthew.R.Wette at jpl.nasa.gov -- I speak for myself, not for JPL.