1,721,090 research outputs found
Control-oriented regularization for linear system identification
No full textIn this paper, we develop a novel theoretical framework for control-oriented identification, based on a Bayesian perspective on modeling. Specifically, we show that closed-loop specifications can be incorporated within the identification procedure as a prior of the model probability distribution via suitable regularization. The corresponding kernel varies according to the additional penalty term and provides a new insight on control-oriented identification. As a secondary contribution, we derive a Bayesian robust control design approach exploiting all the information coming from the above modeling procedure, including the estimate of the uncertainty set The effectiveness of the proposed strategy against state-of-the-art regularized identification is illustrated on a benchmark example for digital control system design
Value-function estimation and uncertainty propagation in Reinforcement Learning: a Koopman operator approach
No full text
Estimating Koopman operators for nonlinear dynamical systems: A nonparametric approach
Full text linkThe Koopman operator provides a linear description of non-linear systems exploiting an embedding into an infinite dimensional space. Dynamic Mode Decomposition and Extended Dynamic Mode Decomposition are amongst the most popular finite dimensional approximations of the Koopman Operator. In this paper we capture their core essence as a dual version of the same problem, embedding them into the Kernel framework. To do so, we leverage the RKHS as a suitable space for learning the Koopman dynamics. Learning from finite length data automatically provides a finite dimensional approximation induced by data. Simulations and comparison with standard procedures are included
Data-Driven Control of Nonlinear Systems: Learning Koopman Operators for Policy Gradient
No full textData-driven control of nonlinear dynamical systems is a largely open problem. In this paper, building upon the theory of Koopman operators and exploiting ideas from policy gradient methods in reinforcement learning, a novel approach for data-driven optimal control of unknown nonlinear dynamical systems is introduced
Kernel-based system identification with manifold regularization: A Bayesian perspective
No full textThis paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by a suitable empirical Bayes approach. The effectiveness of the method in the context of dynamical system identification is evaluated on a simulated linear system and on an experimental switching system setup. (C) 2022 Elsevier Ltd. All rights reserved
Kernel-based system identification with manifold regularization: A Bayesian perspective
No full textThis paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by a suitable empirical Bayes approach. The effectiveness of the method in the context of dynamical system identification is evaluated on a simulated linear system and on an experimental switching system setup
Online semi-parametric learning for inverse dynamics modeling
No full textThis paper presents a semi-parametric algorithm for online learning of a robot inverse dynamics model. It combines the strength of the parametric and non-parametric modeling. The former exploits the rigid body dynamics equation, while the latter exploits a suitable kernel function. We provide an extensive comparison with other methods from the literature using real data from the iCub humanoid robot. In doing so we also compare two different techniques, namely cross validation and marginal likelihood optimization, for estimating the hyperparameters of the kernel function
- …
