—Many controlled systems suffer from unmodeled nonlinear effects that recur periodically over time. Model-free controllers generally cannot compensate these effects, and good physical models for such periodic dynamics are challenging to construct. We investigate nonparametric system identification for periodically recurring nonlinear effects. Within a Gaussian process regression framework, we use a locally periodic covariance function to shape the hypothesis space, which allows for a structured extrapolation that is not possible with more widely used covariance functions. We show that hyperparameter estimation can be performed online by using the maximum aposteriori point estimate, which provides an accuracy comparable to sampling methods as soon as enough data to cover the periodic structure has been collected. It is also shown how the periodic structure can be exploited in the hyperparameter optimization. The predictions obtained from the Gaussian process model are then used in a ...
Edgar D. Klenske, Melanie Nicole Zeilinger, Bernha