Koopman Operator in the Weighted Function Spaces and its Learning for the Estimation of Lyapunov and Zubov Functions

TL;DR

This paper introduces a weighted function space approach to learning the Koopman operator, ensuring contractiveness and bounded errors, to improve the estimation of Lyapunov and Zubov functions for nonlinear stability analysis.

Contribution

It proposes a novel weighted reproducing kernel Hilbert space framework for Koopman operator learning, guaranteeing stability properties and providing probabilistic error bounds.

Findings

The weighted Koopman operator is contractive and has bounded learning errors.

The method accurately estimates Lyapunov functions for stable systems.

Probabilistic error bounds are established for the estimations.

Abstract

The mathematical properties and data-driven learning of the Koopman operator, which represents nonlinear dynamics as a linear mapping on a properly defined functional spaces, have become key problems in nonlinear system identification and control. However, Koopman operators that are approximately learned from snapshot data may not always accurately predict the system evolution on long horizons. In this work, by defining the Koopman operator on a space of weighted continuous functions and learning it on a weighted reproducing kernel Hilbert space, the Koopman operator is guaranteed to be contractive and the accumulation learning error is bounded. The weighting function, assumed to be known a priori, has an exponential decay with the flow or decays exponentially when compensated by an exponential factor. Under such a construction, the Koopman operator learned from data is used to estimate (i) Lyapunov functions for globally asymptotically stable dynamics, and (ii) Zubov-Lyapunov functions that characterize the domain of attraction. For these estimations, probabilistic bounds on the errors are derived.