Search Results for author: Roy S. Smith

Found 36 papers, 0 papers with code

Finite Sample Frequency Domain Identification

no code implementations1 Apr 2024 Anastasios Tsiamis, Mohamed Abdalmoaty, Roy S. Smith, John Lygeros

The error rate is of the order of $\mathcal{O}((d_{\mathrm{u}}+\sqrt{d_{\mathrm{u}}d_{\mathrm{y}}})\sqrt{M/N_{\mathrm{tot}}})$, where $N_{\mathrm{tot}}$ is the total number of samples, $M$ is the number of desired frequencies, and $d_{\mathrm{u}},\, d_{\mathrm{y}}$ are the dimensions of the input and output signals respectively.

valid

Small Noise Analysis of Non-Parametric Closed-Loop Identification

no code implementations23 Mar 2024 Mohamed Abdalmoaty, Roy S. Smith

We revisit the problem of non-parametric closed-loop identification in frequency domain; we give a brief survey of the literature and provide a small noise analysis of the direct, indirect, and joint input-output methods when two independent experiments with identical excitation are used.

Optimal Data-Driven Prediction and Predictive Control using Signal Matrix Models

no code implementations22 Mar 2024 Roy S. Smith, Mohamed Abdalmoaty, Mingzhou Yin

Data-driven control uses a past signal trajectory to characterise the input-output behaviour of a system.

LEMMA

Online Identification of Stochastic Continuous-Time Wiener Models Using Sampled Data

no code implementations9 Mar 2024 Mohamed Abdalmoaty, Efe C. Balta, John Lygeros, Roy S. Smith

It is well known that ignoring the presence of stochastic disturbances in the identification of stochastic Wiener models leads to asymptotically biased estimators.

Frequency-Domain Identification of Discrete-Time Systems using Sum-of-Rational Optimization

no code implementations25 Dec 2023 Mohamed Abdalmoaty, Jared Miller, Mingzhou Yin, Roy S. Smith

We propose a computationally tractable method for the identification of stable canonical discrete-time rational transfer function models, using frequency domain data.

Stochastic Data-Driven Predictive Control: Regularization, Estimation, and Constraint Tightening

no code implementations5 Dec 2023 Mingzhou Yin, Andrea Iannelli, Roy S. Smith

An initial condition estimator is proposed by filtering the measurements with the one-step-ahead stochastic data-driven prediction.

LEMMA

MIMO Grid Impedance Identification of Three-Phase Power Systems: Parametric vs. Nonparametric Approaches

no code implementations29 Apr 2023 Verena Häberle, Linbin Huang, Xiuqiang He, Eduardo Prieto-Araujo, Roy S. Smith, Florian Dörfler

A fast and accurate grid impedance measurement of three-phase power systems is crucial for online assessment of power system stability and adaptive control of grid-connected converters.

Once upon a time step: A closed-loop approach to robust MPC design

no code implementations20 Mar 2023 Anilkumar Parsi, Marcell Bartos, Amber Srivastava, Sebastien Gros, Roy S. Smith

A novel perspective on the design of robust model predictive control (MPC) methods is presented, whereby closed-loop constraint satisfaction is ensured using recursive feasibility of the MPC optimization.

LEMMA Model Predictive Control

Error Bounds for Kernel-Based Linear System Identification with Unknown Hyperparameters

no code implementations17 Mar 2023 Mingzhou Yin, Roy S. Smith

The kernel-based method has been successfully applied in linear system identification using stable kernel designs.

Diagonally Square Root Integrable Kernels in System Identification

no code implementations24 Feb 2023 Mohammad Khosravi, Roy S. Smith

For the stability of a Gaussian process centered at a stable impulse response, we show that the necessary and sufficient condition is the diagonally square root integrability of the corresponding kernel.

Dual adaptive MPC using an exact set-membership reformulation

no code implementations29 Nov 2022 Anilkumar Parsi, Diyou Liu, Andrea Iannelli, Roy S. Smith

Adaptive model predictive control (MPC) methods using set-membership identification to reduce parameter uncertainty are considered in this work.

Model Predictive Control

The Existence and Uniqueness of Solutions for Kernel-Based System Identification

no code implementations17 Apr 2022 Mohammad Khosravi, Roy S. Smith

The consequent estimation problem is well-defined under the central assumption that the convolution operators restricted to the RKHS are continuous linear functionals.

Regret Analysis of Online Gradient Descent-based Iterative Learning Control with Model Mismatch

no code implementations10 Apr 2022 Efe C. Balta, Andrea Iannelli, Roy S. Smith, John Lygeros

In Iterative Learning Control (ILC), a sequence of feedforward control actions is generated at each iteration on the basis of partial model knowledge and past measurements with the goal of steering the system toward a desired reference trajectory.

Computationally efficient robust MPC using optimized constraint tightening

no code implementations5 Apr 2022 Anilkumar Parsi, Panagiotis Anagnostaras, Andrea Iannelli, Roy S. Smith

A robust model predictive control (MPC) method is presented for linear, time-invariant systems affected by bounded additive disturbances.

Model Predictive Control

Scalable tube model predictive control of uncertain linear systems using ellipsoidal sets

no code implementations5 Apr 2022 Anilkumar Parsi, Andrea Iannelli, Roy S. Smith

This contrasts with most existing tube MPC strategies using polytopic sets in the state tube, which are difficult to design and whose complexity grows combinatorially with the system order.

Model Predictive Control

Kernel-Based Identification of Local Limit Cycle Dynamics with Linear Periodically Parameter-Varying Models

no code implementations30 Mar 2022 Defne E. Ozan, Mingzhou Yin, Andrea Iannelli, Roy S. Smith

Limit cycle oscillations are phenomena arising in nonlinear dynamical systems and characterized by periodic, locally-stable, and self-sustained state trajectories.

Infinite-Dimensional Sparse Learning in Linear System Identification

no code implementations28 Mar 2022 Mingzhou Yin, Mehmet Tolga Akan, Andrea Iannelli, Roy S. Smith

Atomic norm regularization decomposes the transfer function into first-order atomic models and solves a group lasso problem that selects a sparse set of poles and identifies the corresponding coefficients.

Sparse Learning

Data-Driven Prediction with Stochastic Data: Confidence Regions and Minimum Mean-Squared Error Estimates

no code implementations8 Nov 2021 Mingzhou Yin, Andrea Iannelli, Roy S. Smith

In this paper, confidence regions are provided for these stochastic predictors based on the prediction error distribution.

valid

Kernel-based Impulse Response Identification with Side-Information on Steady-State Gain

no code implementations31 Oct 2021 Mohammad Khosravi, Roy S. Smith

The objective function of this optimization problem is the empirical loss regularized with the norm of RKHS, and the constraint is considered for enforcing the integration of the steady-state gain side-information.

Regularized Identification with Internal Positivity Side-Information

no code implementations31 Oct 2021 Mohammad Khosravi, Roy S. Smith

Utilizing these conditions, the impulse response estimation problem is formulated as a constrained optimization in a reproducing kernel Hilbert space equipped with a stable kernel, and suitable constraints are imposed to encode the internal positivity side-information.

Physics-informed linear regression is competitive with two Machine Learning methods in residential building MPC

no code implementations29 Oct 2021 Felix Bünning, Benjamin Huber, Adrian Schalbetter, Ahmed Aboudonia, Mathias Hudoba de Badyn, Philipp Heer, Roy S. Smith, John Lygeros

However, we also see that the physics-informed ARMAX models have a lower computational burden, and a superior sample efficiency compared to the Machine Learning based models.

BIG-bench Machine Learning regression

A distributed framework for linear adaptive MPC

no code implementations13 Sep 2021 Anilkumar Parsi, Ahmed Aboudonia, Andrea Iannelli, John Lygeros, Roy S. Smith

Adaptive model predictive control (MPC) robustly ensures safety while reducing uncertainty during operation.

Model Predictive Control

Design of input for data-driven simulation with Hankel and Page matrices

no code implementations10 Sep 2021 Andrea Iannelli, Mingzhou Yin, Roy S. Smith

The paper deals with the problem of designing informative input trajectories for data-driven simulation.

Robust Adaptive Model Predictive Control of Quadrotors

no code implementations26 Feb 2021 Alexandre Didier, Anilkumar Parsi, Jeremy Coulson, Roy S. Smith

To the best of our knowledge this is the first time that RAMPC has been applied in practice using a state space formulation.

Model Predictive Control

The Balanced Mode Decomposition Algorithm for Data-Driven LPV Low-Order Models of Aeroservoelastic Systems

no code implementations7 Feb 2021 Andrea Iannelli, Urban Fasel, Roy S. Smith

The algorithm formulation hinges on the idea of replacing the orthogonal projection onto the Proper Orthogonal Decomposition modes, used in Dynamic Mode Decomposition-based approaches, with a balancing oblique projection constructed entirely from data.

Model Predictive Control

Parameter Identification for Digital Fabrication: A Gaussian Process Learning Approach

no code implementations20 Dec 2020 Yvonne R. Stürz, Mohammad Khosravi, Roy S. Smith

This is beneficial since measurements of the cable net form on the construction site are very expensive.

Gaussian Processes

Experiment design for impulse response identification with signal matrix models

no code implementations15 Dec 2020 Andrea Iannelli, Mingzhou Yin, Roy S. Smith

This paper formulates an input design approach for truncated infinite impulse response identification in the context of implicit model representations recently used as basis for data-driven simulation and control approaches.

On Low-Rank Hankel Matrix Denoising

no code implementations14 Dec 2020 Mingzhou Yin, Roy S. Smith

The low-complexity assumption in linear systems can often be expressed as rank deficiency in data matrices with generalized Hankel structure.

Denoising

Extended Full Block S-Procedure for Distributed Control of Interconnected Systems

no code implementations11 Dec 2020 Giulia De Pasquale, Yvonne R. Sturz, Maria Elena Valcher, Roy S. Smith

This paper proposes a novel method for distributed controller synthesis of homogeneous interconnected systems consisting of identical subsystems.

Maximum Likelihood Signal Matrix Model for Data-Driven Predictive Control

no code implementations8 Dec 2020 Mingzhou Yin, Andrea Iannelli, Roy S. Smith

The paper presents a data-driven predictive control framework based on an implicit input-output mapping derived directly from the signal matrix of collected data.

Model Predictive Control

Safety-Aware Cascade Controller Tuning Using Constrained Bayesian Optimization

no code implementations28 Oct 2020 Christopher König, Mohammad Khosravi, Markus Maier, Roy S. Smith, Alisa Rupenyan, John Lygeros

This paper presents an automated, model-free, data-driven method for the safe tuning of PID cascade controller gains based on Bayesian optimization.

Bayesian Optimization Gaussian Processes

Robust MPC with data-driven demand forecasting for frequency regulation with heat pumps

no code implementations15 Sep 2020 Felix Bünning, Joseph Warrington, Philipp Heer, Roy S. Smith, John Lygeros

By combining a control scheme based on Robust Model Predictive Control, with affine policies, and heating demand forecasting based on Artificial Neural Networks with online correction methods, we offer frequency regulation reserves and maintain user comfort with a system comprising a heat pump and buffer storage.

Model Predictive Control

Subspace Identification of Linear Time-Periodic Systems with Periodic Inputs

no code implementations6 Jun 2020 Mingzhou Yin, Andrea Iannelli, Roy S. Smith

The response is estimated with an ensemble of input-output data with periodic inputs.

Structured exploration in the finite horizon linear quadratic dual control problem

no code implementations31 Oct 2019 Andrea Iannelli, Mohammad Khosravi, Roy S. Smith

This paper presents a novel approach to synthesize dual controllers for unknown linear time-invariant systems with the tasks of optimizing a quadratic cost while reducing the uncertainty.

Efficient Exploration

Cannot find the paper you are looking for? You can Submit a new open access paper.