Roger Labbe
8e2bb6856d
Fixed error in latex
...
It rendered fine in the notebook, but \text{foo_bar} was causing
an error in the pdf generation. Switching to \mathtt{foo\_bar}
mostly fixes it (though in the notebook it renders as -, not _.
Oye.
2016-01-09 09:10:56 -08:00
Roger Labbe
800ea6c189
Cleaned up equations for UKF.
...
I was using a bunch of variable names that weren't consistent
with the rest of the book (but perhaps are more consistent with
the literature). It just made everything more challenging than
it needed to be, so instead of \mu and \sigma (e.g.) I use
\bar x and \bar P.
I also am in the middle of rewriting some sections for clarity,
but that work is not completed.
2016-01-09 08:52:03 -08:00
Roger Labbe
b6d5052f45
Copy edits.
2016-01-09 08:51:24 -08:00
Roger Labbe
eeab89c16c
Copy edits.
2016-01-09 08:51:03 -08:00
Roger Labbe
6b9b97927d
Redid Discrete Bayes animations.
...
I added titles so you can see where in the processing the
graph is.
This surfaced a bug in matplotlib 1.5 (#5399 ) so I added
some code to work around that.
2016-01-09 08:49:29 -08:00
Roger Labbe
65a77d0af2
Commited before I saved the file!
2016-01-09 08:20:53 -08:00
Roger Labbe
8bab476b31
Edited so I can report bug to matplotlib.
2016-01-09 08:20:28 -08:00
Roger Labbe
cb45bf236e
Copy edits
2016-01-05 15:45:27 -08:00
Roger Labbe
6a621ca59e
Merge pull request #76 from wspr/wspr
...
03-gaussians copyedit
2016-01-05 06:44:28 -08:00
Will Robertson
615501e899
03-gaussians copyedit
...
* integrals should have "du", etc. (right??)
* $\mathit{VAR}$ gives better spacing than $VAR$
* I find "within $1\sigma$" to be ambiguous; "within $\pm1sigma$" much more clear
* Changed dashes in text from hyphens to em-dashes
* Changed (some) straight quotation marks to curly ones (and removed single quotes in favour of double ones)
* Improved "°C" to just use a text symbol
* Assorted random typos, etc., but avoided rewording anything except for one particularly difficult sentence
2016-01-06 00:26:48 +10:30
Roger Labbe
e52e2ca206
Copy edits.
2016-01-03 16:13:40 -08:00
Roger Labbe
8a01161ea4
Streamlined description of EKF linearization
2016-01-03 11:32:22 -08:00
Roger Labbe
331d8cf880
Minor copy editing.
2016-01-03 11:32:07 -08:00
Roger Labbe
462ac1bb5c
Fixed obsolete import, hard code of filename
2016-01-03 10:43:45 -08:00
Roger Labbe
122fe27680
changed from \bar to \overbar for prediction.
...
More visible, and works better with things like matrices.
2016-01-02 18:10:15 -08:00
Roger Labbe
ab33584a0c
Minor copy edit and code clean up.
...
Biggest change was changing the variable for noise to use
either std or var.
2016-01-02 18:08:43 -08:00
Roger Labbe
09406a6419
Added explanation for system dynamics matrix.
2016-01-02 18:07:00 -08:00
Roger Labbe
fee08bfc7b
Moved iterative ILD to supporting notebooks
...
Most people just won't want to read this material.
2016-01-01 15:43:50 -08:00
Roger Labbe
47cce41efb
Moved Taylor Series to supporting notebooks.
2016-01-01 15:28:55 -08:00
Roger Labbe
b2a3ce2a28
Major rewrite of matrix exponential section.
...
I was conflating thematrix exponential with Taylor series, making
the whole thing quite confusing.
There was also significant duplication of content, such as presenting
the calculus form of Newton's equations multiple times in slightly
different ways. I've made it all hang together now.
2016-01-01 15:18:35 -08:00
Roger Labbe
2610c4dc85
Merge branch 'master' of https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python
2015-12-27 08:13:14 -08:00
Roger Labbe
5999539d56
Merge pull request #75 from rsvp/devCode
...
code directory: Add __init__.py for access as package
2015-12-27 08:12:52 -08:00
Adriano rsvp.github.com
6444d77818
code directory: Add __init__.py for access as package
...
Useful for absolute import.
2015-12-26 17:24:30 -08:00
Roger Labbe
63c398fc51
changed \bar to \overline
2015-12-20 12:42:30 -08:00
Roger Labbe
3445153b7c
Latex cleanup.
2015-12-20 12:23:09 -08:00
Roger Labbe
5ffd42a25d
Fixed FilterPy versioning check error.
...
The code only worked for the third version number. If the first
or second version numbers were too small it didn't detect it.
2015-12-20 10:08:31 -08:00
Roger Labbe
033dbea735
Copy edits.
2015-12-20 10:08:15 -08:00
Roger Labbe
108a3f1a4d
Copy edits.
2015-12-20 09:00:26 -08:00
Roger Labbe
672e4bb3c5
Updated code listings.
...
Note sure I like this appendix, but as long as it exists it should
be up to date.
2015-12-20 08:59:58 -08:00
Roger Labbe
0e86469fac
PDF build scripts build 6x9 book separately.
2015-12-20 08:59:11 -08:00
Roger Labbe
05caf4f1bb
Copy edits.
2015-12-20 08:44:28 -08:00
Roger Labbe
9577007cc6
Deleted instructions for Sagemath install.
...
It is just too complicated for this book.
2015-12-20 08:34:42 -08:00
Roger Labbe
dd1bb52baa
Improved description of KFs.
2015-12-20 08:32:06 -08:00
Roger Labbe
0236af896d
Added text for computation of velocities.
...
I explained how the covariances are computed by FPF',
and then how the velocity is computed using the kalman gain.
2015-12-19 21:12:41 -08:00
Roger Labbe
77ba700e39
Revised particle filter chapter.
...
Pretty happy with it now. Needs copy editing, and probably an
easier introduction to convey the basic idea. Moved from a class
based approach to a procedural approach, and I like that very much.
2015-12-19 16:18:21 -08:00
Roger Labbe
3c270d0c87
Fixed latex typo.
2015-12-19 16:17:40 -08:00
Roger Labbe
ba553f2068
Fixes for GitHub issue #65 .
...
Minor wording changes to clarify meaning.
2015-12-15 16:05:01 -08:00
Roger Labbe
e12352c520
Fixed equation for total probability theorem.
...
I was using the wrong subscript (i instead of j) in the sum.
2015-12-15 15:45:26 -08:00
Roger Labbe
04ce859d01
Minor copy editing.
2015-12-13 15:16:33 -08:00
Roger Labbe
7905298583
Improved description of state-space methods.
2015-12-13 15:16:00 -08:00
Roger Labbe
26125c538f
Added example of KF without using FilterPy.
2015-12-13 15:15:27 -08:00
Roger Labbe
6a05d8edd7
Fixed spelling mistake.
2015-12-13 11:01:15 -08:00
Roger Labbe
01dab63459
Added reference to Grewal2008 for observability.
2015-12-13 10:53:45 -08:00
Roger Labbe
a1a1753c18
Fixed description of fixed-point filter.
...
I mistakenly stated that it re-estimated the points between times
j and k, when of course it recomputes only a single fixed point.
2015-12-10 20:21:13 -08:00
Roger Labbe
9ece51233b
Minor copy edits.
2015-12-10 19:31:41 -08:00
Roger Labbe
0ca7aa1e84
Reran cells so they are executed in sequence.
2015-12-10 19:31:16 -08:00
Roger Labbe
5794485341
Added animation to PF chapter.
2015-12-10 18:13:41 -08:00
Roger Labbe
41d8d246e0
Extensive copy editing.
2015-12-09 06:31:14 -08:00
Roger Labbe
7b218e7243
Fixed typos.
2015-12-09 06:30:42 -08:00
Roger Labbe
ef28d71938
Minor rewordings.
2015-11-29 20:46:49 -08:00