This package implements a family of Robust Kalman filters.
Each robust Kalman filter is selected by fixing the paramter tau (real value between 0 and 1).
The robustness of the filter is tuned by the tolerance c.

The robust robust Kalman is designed knowing that the true model belongs to
a ball about the nominal one. The models in that ball are such that
the Tau-divergence between them and the nominal model is less than
tolerance c.

This package also contains an example demonstrating its practical application.

References:
M. Zorzi. "Robust Kalman Filtering under Model Perturbations". Submitted.
M. Zorzi. "On the Robustness of the Bayes and Wiener Estimators under Model Uncertainty".