ssmtoybox 0.1.1a0

Local filters based on Bayesian quadrature

SSM Toybox

Python 3 implementation of the nonlinear sigma-point filters based on Bayesian quadrature, such as

• Gaussian Process Quadrature Kalman Filter [1]
• Student's t-Process Quadrature Kalman Filter [2]

Included are also the well-known classical nonlinear Kalman filters such as:

• Extended Kalman Filter
• Unscented Kalman Filter
• Cubature Kalman Filter
• Gauss-Hermite Kalman Filter

Build documentation

cd docs
sphinx-apidoc -o ./ ../ssmtoybox ../ssmtoybox/tests
make html

Why toybox?

Because 'toolbox' sounds too serious :-).

References

[1]: [DOI | PDF] Prüher, J. and Straka, O. Gaussian Process Quadrature Moment Transform, IEEE Transactions on Automatic Control, 2017

[2]: [DOI | PDF] Prüher, J.; Tronarp, F.; Karvonen, T.; Särkkä, S. and Straka, O. Student-t Process Quadratures for Filtering of Non-linear Systems with Heavy-tailed Noise, 20th International Conference on Information Fusion (Fusion), 1-8, 2017