Searcharxiv⌕ Search

arXiv subjects

Dimitrios Moustroufis

Publications and source records attributed to Dimitrios Moustroufis.

2 recordsLinked to original sources

DiSKO: Distribution Steering for Nonlinear Systems using the Koopman Operator

This paper studies the problem of data-driven covariance steering for stochastic, nonlinear dynamical systems. We utilize Koopman operator theory to learn, directly from data, an approximate linear representation of the nonlinear dynamics in a lifted space of observable functions of the state. By exploiting this linear model, we design controllers for the mean and covariance of the underlying nonlinear system by utilizing linear distribution steering algorithms. To account for approximation errors in the learned dynamics and the ambiguity in the disturbance distribution induced in the lifted space, we formulate the problem as a distributionally robust covariance steering problem. Furthermore, we leverage a deep neural network (DNN) to simultaneously learn the Koopman observables and the linear model parameters, thus eliminating the need for manual observables basis selection while also reducing the dimensionality of the lifted representation. Additionally, we can promote desirable model properties, such as stability, during DNN training. We demonstrate the effectiveness of the proposed framework via numerical simulations.

eess.SY↗

Data-Driven Covariance Steering with Output Feedback

This paper addresses output-feedback covariance steering for stochastic discrete-time linear time-invariant systems with unknown dynamics. We construct a controllable non-minimal state representation from past inputs and outputs, allowing the problem to be formulated in a standard state-feedback setting. The induced disturbance, however, is temporally correlated, requiring propagation of the state-disturbance cross-covariance. Using persistently exciting offline data, we employ an indirect approach for mean steering and a direct approach for covariance steering. The indirect formulation requires estimation of the mean dynamics, whereas the direct formulation requires estimation of the noise realization. We develop estimation methods suitable for temporally correlated noise and formulate the resulting covariance steering problem as a convex semidefinite program. Numerical simulations demonstrate the effectiveness of the proposed framework.

eess.SY↗