Code

libDirectional

libDirectional (Kurz et al., 2019) is a MATLAB library implementing most of my methods and a number of methods found in literature from the field of Directional Statistics. It is freely available on Github and licensed under the GPL v3.

https://github.com/libDirectional/libDirectional

NonlinEst

NonlinEst is a Python library for nonlinear estimation that contains a variety of nonlinear filters such as Extended Kalman filter (EKF), Unscented Kalman filter (UKF), Cubature Kalman filter (CFK), Ensemble Kalman filter (EnKF), various particle filters etc. It is available on Github under an Apache-2.0 license.

https://github.com/boschresearch/nonlinest

Other Projects

• https://github.com/gerhardkurz/VideoFrameworkMATLAB
  • a framework to create videos with MATLAB
  • uses ffmpeg to encode the video
• https://github.com/gerhardkurz/MatlabUML
  • a tool to automatically generate UML class diagrams from MATLAB code
  • uses graphviz to generate the diagram
• https://github.com/gerhardkurz/FastEM
  • an efficient implementation of the Expectation Maximization algorithm
  • supports weighted samples
  • provides a mex-file that can be called easily from MATLAB
• https://github.com/gerhardkurz/TransportAuctionMex
  • auction algorithm for transport problems
  • MATLAB-wrapper for the C++ implementation by J. D. Walsh
• https://github.com/gerhardkurz/TessellateS3
  • tessellatation of unit sphere in 4D using tetrahedral/octahedral subdivision
  • MATLAB-wrapper for the C implementation by Jared Glover in libBingham

References

1. Kurz, G., Gilitschenski, I., Pfaff, F., Drude, L., Hanebeck, U., Haeb-Umbach, R., & Siegwart, R. (2019). Directional Statistics and Filtering Using libDirectional. Journal of Statistical Software, Articles, 89(4), 1–31. https://doi.org/10.18637/jss.v089.i04