RcppKalman

A C++ port using RcppArmadillo of the EKF/UKF toolbox for Matlab

Why?

R has a number of existing Kalman filter packages which are all very noteworthy in their own right. A comparison was provided by this JSS paper from 2011.

Yet I had a need for something both simple and fast at the C++ level.

The EKF/UKF toolbox for Matlab proved to be a wonderful source of excellent code that was well documented (see [this 130 page pdf manual]http://becs.aalto.fi/en/research/bayes/ekfukf/documentation.pdf()), under a suitable license and covering both simple examples as well as promising extensions.

Example

The code

library(RcppKalman)
demo(kf_sine_demo)

produces this chart

RcppKalman demo chart
which reproduces Figures 3.1 and 3.2 from the third chapter (on nonlinear state-space estimation) of the UKF/EKF manual.

Where

The package is hosted in this GitHub repository and as of this writing not yet on CRAN.

License

GPL (>= 2) just like R, Rcpp, RcppArmadillo and EKF/UKF.

Status

The package is still pretty young and incomplete.

Author

Dirk Eddelbuettel for this package.
Simo Särkkä, Jouni Hartikainen, and Arno Solin for EKF/UKF.

Initially created: Sun Apr 27 07:34:20 CDT 2014
Last modified: Tue Jan 06 21:35:35 CST 2015