The book kind of side stepped the Python implementation for awhile.
I put the implementation much earlier to give reader familiarity with
the equations earlier. I moved the ooptional walkthrough to the end of
the text, but it needs extensive editting as the code is quite obsolete.
Added text that represented the multivariate KF equations with scalars
for the univariate case to illustrate how simple the equations actually
are. Then to complete the section I proved the equivelence of the last
chapter's equations with this one using algebra.
Added exercises comparing 1D filter using Gaussians vs 1D filter
using KalmanFilter class, and comparing a 1D filter vs a filter
incorporating the unobserved derivative of x.
I originally used sense/update as my function names. But all the
literature uses update/predict, so I am normalizing on that.
I also wrote a test to ensure the update/predict functions in the
1d kalman filter chapter gives the same output as the full blown
kalman filter. Of course, they do. This is in the 1d_kf_compare
notebook. I will add this as an exercise in the multidimensional KF
chapter.
Added text in book about multivariate Gaussian. Wrote tests to confirm
that my multivariate function yields the same results as numpy's
version, and made everything compliant with py.test.
I switched my code over to this nomenclature a while ago, but have not
updated the book for the switch.
As an aside, the code for the Kalman filter class in theb book is now
quite dated. I need to incorporate filterpy library more fully.
Text is still not organized well, but I got the unscented transform fully documented. Probably needs more citations to the original article, and some revisions - I think there is some duplication of info.
Started developing EKF chapter. Stalled - not sure what is the best
initial example to develop. Air drag seems unnecessarily 'mathy',
but then you need diff eq to compute Jacobians.
