Change search
Link to record
Permanent link

Direct link
BETA
Alternative names
Publications (10 of 144) Show all publications
Pan, S., González, R. A., Welsh, J. S. & Rojas, C. R. (2020). Consistency analysis of the Simplified Refined Instrumental Variable method for Continuous-time systems. Automatica, 113, Article ID 108767.
Open this publication in new window or tab >>Consistency analysis of the Simplified Refined Instrumental Variable method for Continuous-time systems
2020 (English)In: Automatica, ISSN 0005-1098, E-ISSN 1873-2836, Vol. 113, article id 108767Article in journal (Refereed) Published
Abstract [en]

In this paper, we analyse the consistency of the Simplified Refined Instrumental Variable method for Continuous-time systems (SRIVC). It is well known that the intersample behaviour of the input signal influences the quality and accuracy of the results when estimating and simulating continuous-time models. Here, we present a comprehensive analysis on the consistency of the SRIVC estimator while taking into account the intersample behaviour of the input signal. The main result of the paper shows that, under some mild conditions, the SRIVC estimator is generically consistent. We also describe some conditions when consistency is not achieved, which is important from a practical standpoint. The theoretical results are supported by simulation examples.

Place, publisher, year, edition, pages
Elsevier Ltd, 2020
Keywords
Consistency, Continuous-time systems, Identification, Instrumental variable method, Control systems, Electrical engineering, Identification (control systems), Comprehensive analysis, Consistency analysis, Continuous time models, Instrumental variable methods, Refined instrumental variables, Simulation example, Continuous time systems
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-272258 (URN)10.1016/j.automatica.2019.108767 (DOI)000514216600002 ()2-s2.0-85076691878 (Scopus ID)
Note

QC 20200427

Available from: 2020-04-27 Created: 2020-04-27 Last updated: 2020-04-27Bibliographically approved
Rojas, C. R. & Müller, M. I. (2019). Algorithms for data-driven H∞-norm estimation. In: Carlo Novara and Simone Formentin (Ed.), DATA-DRIVEN FILTER AND CONTROL DESIGN: Methods and applications (pp. 145-163). IET Digital Library
Open this publication in new window or tab >>Algorithms for data-driven H-norm estimation
2019 (English)In: DATA-DRIVEN FILTER AND CONTROL DESIGN: Methods and applications / [ed] Carlo Novara and Simone Formentin, IET Digital Library, 2019, p. 145-163Chapter in book (Refereed)
Abstract [en]

In this chapter, the problem of estimating in a model-free manner the H norm of a linear dynamic system is discussed at a tutorial level. Two recently developed methods for addressing this problem are presented, namely the power iterations method and a class of multi-armed bandit (MAB) algorithms. Due to reasons of space, many details are omitted, but references are provided to complement this exposition.

Place, publisher, year, edition, pages
IET Digital Library, 2019
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-257879 (URN)10.1049/PBCE123E (DOI)9781785617126 (ISBN)
Note

QC 20190909

Available from: 2019-09-06 Created: 2019-09-06 Last updated: 2019-09-09Bibliographically approved
Müller, M. I. & Rojas, C. R. (2019). Gain estimation of linear dynamical systems using Thompson Sampling. In: Kamalika Chaudhuri, Masashi Sugiyama (Ed.), Proceedings of Machine Learning Research: . Paper presented at The 22nd International Conference on Artificial Intelligence and Statistics (AISTATS) (pp. 1535-1543). , 89
Open this publication in new window or tab >>Gain estimation of linear dynamical systems using Thompson Sampling
2019 (English)In: Proceedings of Machine Learning Research / [ed] Kamalika Chaudhuri, Masashi Sugiyama, 2019, Vol. 89, p. 1535-1543Conference paper, Published paper (Refereed)
National Category
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:kth:diva-256045 (URN)
Conference
The 22nd International Conference on Artificial Intelligence and Statistics (AISTATS)
Note

QC 20190820

Available from: 2019-08-16 Created: 2019-08-16 Last updated: 2019-08-20Bibliographically approved
Müller, M. I. & Rojas, C. R. (2019). Gain estimation of linear dynamical systems using Thompson Sampling. In: Chaudhuri, K Sugiyama, M (Ed.), 22ND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 89: . Paper presented at 22nd International Conference on Artificial Intelligence and Statistics (AISTATS), APR 16-18, 2019, Naha, JAPAN. MICROTOME PUBLISHING
Open this publication in new window or tab >>Gain estimation of linear dynamical systems using Thompson Sampling
2019 (English)In: 22ND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 89 / [ed] Chaudhuri, K Sugiyama, M, MICROTOME PUBLISHING , 2019Conference paper, Published paper (Refereed)
Abstract [en]

We present the gain estimation problem for linear dynamical systems as a multi-armed bandit. This is particularly a very important engineering problem in control design, where performance guarantees are carted in terms of the largest gain of the frequency response of the system. The dynamical system is unknown and only noisy input-output data is available. In a more general setup, the noise perturbing the data is non-white and the variance at each frequency band is unknown, resulting in a two-dimensional Gaussian bandit model with unknown mean and scaled-identity covariance matrix. This model corresponds to a two-parameter exponential family. Within a bandit framework, the set of means is given by the frequency response of the system and, unlike traditional bandit problems, the goal here is to maximize the probability of choosing the arm drawing samples with the highest norm of its mean. A problem-dependent lower bound for the expected cumulative regret is derived and a matching upper bound is obtained for a Thompson-Sampling algorithm under a uniform prior over the variances and the two-dimensional means.

Place, publisher, year, edition, pages
MICROTOME PUBLISHING, 2019
Series
Proceedings of Machine Learning Research, ISSN 2640-3498 ; 89
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:kth:diva-269510 (URN)000509687901060 ()
Conference
22nd International Conference on Artificial Intelligence and Statistics (AISTATS), APR 16-18, 2019, Naha, JAPAN
Note

QC 20200309

Available from: 2020-03-09 Created: 2020-03-09 Last updated: 2020-03-09Bibliographically approved
Valenzuela, P. E., Schön, T. B. & Rojas, C. R. (2019). On model order priors for Bayesian identification of SISO linear systems. International Journal of Control, 92(7), 1645-1661
Open this publication in new window or tab >>On model order priors for Bayesian identification of SISO linear systems
2019 (English)In: International Journal of Control, ISSN 0020-7179, E-ISSN 1366-5820, Vol. 92, no 7, p. 1645-1661Article in journal (Refereed) Published
Abstract [en]

A method for the identification of single input single output linear systems is presented. The method employs a Bayesian approach to compute the posterior distribution of the model parameters given the data-set. Since this distribution is often unavailable in closed form, a Metropolis Hastings algorithm is implemented to draw samples from it. To implement the sampler, the inclusion of prior information regarding the model order of the identified system is discussed. As one of the main contributions of this work, a prior over the Hankel singular values of the model is imposed. Numerical examples illustrate the method.

Place, publisher, year, edition, pages
Taylor & Francis, 2019
Keywords
System identification, Bayesian estimation, Metropolis Hastings sampler, model order prior
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-255303 (URN)10.1080/00207179.2017.1406147 (DOI)000472558100017 ()2-s2.0-85067805258 (Scopus ID)
Note

QC 20190730

Available from: 2019-07-30 Created: 2019-07-30 Last updated: 2019-07-30Bibliographically approved
Galrinho, M., Rojas, C. R. & Hjalmarsson, H. (2019). Parametric Identification Using Weighted Null-Space Fitting. IEEE Transactions on Automatic Control, 64(7), 2798-2813
Open this publication in new window or tab >>Parametric Identification Using Weighted Null-Space Fitting
2019 (English)In: IEEE Transactions on Automatic Control, ISSN 0018-9286, E-ISSN 1558-2523, Vol. 64, no 7, p. 2798-2813Article in journal (Refereed) Published
Abstract [en]

In identification of dynamical systems, the prediction error method with a quadratic cost function provides asymptotically efficient estimates under Gaussian noise, but in general it requires solving a nonconvex optimization problem, which may imply convergence to nonglobal minima. An alternative class of methods uses a nonparametric model as intermediate step to obtain the model of interest. Weighted null-space fitting (WNSF) belongs to this class, starting with the estimate of a nonparametric ARX model with least squares. Then, the reduction to a parametric model is a multistep procedure where each step consists of the solution of a quadratic optimization problem, which can be obtained with weighted least squares. The method is suitable for both open- and closed-loop data, and can be applied to many common parametric model structures, including output-error, ARMAX, and Box-Jenkins. The price to pay is the increase of dimensionality in the nonparametric model, which needs to tend to infinity as function of the sample size for certain asymptotic statistical properties to hold. In this paper, we conduct a rigorous analysis of these properties: namely, consistency, and asymptotic efficiency. Also, we perform a simulation study illustrating the performance of WNSF and identify scenarios where it can be particularly advantageous compared with state-of-the-art methods.

Place, publisher, year, edition, pages
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2019
Keywords
Least squares, system identification
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-255416 (URN)10.1109/TAC.2018.2877673 (DOI)000473489700011 ()2-s2.0-85055726363 (Scopus ID)
Note

QC 20190815

Available from: 2019-08-15 Created: 2019-08-15 Last updated: 2019-10-15Bibliographically approved
Müller, M. I. & Rojas, C. R. (2019). Risk-Coherent H∞-optimal Filter Design Under Model Uncertainty with Applications to MISO Control. In: 2019 18th European Control Conference, ECC 2019: . Paper presented at 18th European Control Conference, ECC 2019; Naples; Italy; 25 June 2019 through 28 June 2019 (pp. 1461-1466). Institute of Electrical and Electronics Engineers (IEEE), Article ID 8795947.
Open this publication in new window or tab >>Risk-Coherent H-optimal Filter Design Under Model Uncertainty with Applications to MISO Control
2019 (English)In: 2019 18th European Control Conference, ECC 2019, Institute of Electrical and Electronics Engineers (IEEE), 2019, p. 1461-1466, article id 8795947Conference paper, Published paper (Refereed)
Abstract [en]

This work presents a framework to address the problem of designing discrete-time LTI (linear and time-invariant) multiple-input and multiple-output (MIMO) filters, aiming to optimize the performance of a system when model uncertainty is considered. Additionally, we present an interesting application to control design for disturbance rejection under model uncertainty. To account for this uncertainty we employ coherent measures of risk, which are a family of measures in theory of risk. We particularly discuss which measures are suitable by comparing the conditional value-at-risk (CVaR) to other three common designs. Using a scenario approach, we derive a convex optimization problem based on linear matrix inequalities (LMIs), whose solution minimizes the risk of falling into poor mathcal{H}{ infty} performance. Finally, we present an application to multiple-input and single-output (MISO) control design under model uncertainty in the auto-covariance function of the output noise, comparing approaches minimizing different notions of risk.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2019
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-256077 (URN)10.23919/ECC.2019.8795947 (DOI)000490488301079 ()2-s2.0-85071597588 (Scopus ID)9783907144008 (ISBN)
Conference
18th European Control Conference, ECC 2019; Naples; Italy; 25 June 2019 through 28 June 2019
Note

QC 20190827

Available from: 2019-08-19 Created: 2019-08-19 Last updated: 2019-11-14Bibliographically approved
Sadeghi, M., Rojas, C. R. & Wahlberg, B. (2018). A Branch and Bound Approach to System Identification based on Fixed-rank Hankel Matrix Optimization. IFAC-PapersOnLine, 51(15), 96-101
Open this publication in new window or tab >>A Branch and Bound Approach to System Identification based on Fixed-rank Hankel Matrix Optimization
2018 (English)In: IFAC-PapersOnLine, E-ISSN 2405-8963, Vol. 51, no 15, p. 96-101Article in journal (Refereed) Published
Abstract [en]

We consider identification of linear systems with a certain order from a set of noisy input-output observations. We utilize the fact that the system order corresponds to the rank of the Hankel matrix associated with the system impulse response. Then, the system identification problem is formulated as the minimization of the output error subject to a rank constraint on a Hankel matrix. As this problem is non-convex, we propose a branch and bound (BB) solver, which is a powerful tool for solving non-convex problems to optimality. The main ingredients of the proposed BB method are a convex relaxation problem and a local minimizer of the original non-convex problem. We illustrate the promising performance of the proposed scheme in a system identification problem. The results demonstrate the higher accuracy and stability of our method in estimating the true system compared to the standard output error (OE) algorithm.

Place, publisher, year, edition, pages
Elsevier B.V., 2018
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-247401 (URN)10.1016/j.ifacol.2018.09.097 (DOI)000446599200018 ()2-s2.0-85054351520 (Scopus ID)
Note

QC 20190322

Available from: 2019-03-22 Created: 2019-03-22 Last updated: 2020-03-04Bibliographically approved
González, R. A. & Rojas, C. R. (2018). A fully Bayesian approach to kernel-based regularization for impulse response estimation⁎. IFAC-PapersOnLine, 51(15), 186-191
Open this publication in new window or tab >>A fully Bayesian approach to kernel-based regularization for impulse response estimation⁎
2018 (English)In: IFAC-PapersOnLine, E-ISSN 2405-8963, Vol. 51, no 15, p. 186-191Article in journal (Refereed) Published
Abstract [en]

Kernel-based regularization has recently been shown to be a successful method for impulse response estimation. This technique usually requires choosing a vector of hyper-parameters in order to form an appropriate regularization matrix. In this paper, we develop an alternative way to obtain kernel-based regularization estimates by Bayesian model mixing. This new approach is tested against state-of-the-art methods for hyperparameter tuning in regularized FIR estimation, with favorable results in many cases.

Place, publisher, year, edition, pages
Elsevier B.V., 2018
Keywords
Bayesian estimation, kernel-based regularization, Linear system identification, model mixing, Impulse response, Linear systems, Mixing, Bayesian estimations, Bayesian model, Fully bayesian approaches, Hyper-parameter, New approaches, State-of-the-art methods, Bayesian networks
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-247500 (URN)10.1016/j.ifacol.2018.09.123 (DOI)000446599200033 ()2-s2.0-85054355409 (Scopus ID)
Available from: 2019-04-03 Created: 2019-04-03 Last updated: 2019-04-03Bibliographically approved
Müller, M. I. & Rojas, C. R. (2018). A Markov Chain Approach to Compute the ℓ2-gain of Nonlinear Systems. In: : . Paper presented at 18th IFAC Symposium on System Identification (pp. 84-89). Elsevier B.V., 51(15)
Open this publication in new window or tab >>A Markov Chain Approach to Compute the ℓ2-gain of Nonlinear Systems
2018 (English)Conference paper, Published paper (Refereed)
Abstract [en]

In this work the problem of computing the maximum gain of non-linear systems, also known as its ℓ2-gain, from input-output data is studied. From an input design perspective, this problem reduces to find an optimal input sequence, of bounded norm, maximizing the norm gain of the output, where our target estimation corresponds to the ratio of these quantities. The novelty of this approach lies on the fact that the input signal is a realization of a stationary process with finite memory whose range is a finite set of values. Based on recent developents on input design for nonlinear systems, our approach leads to a linear program whose optimal cost gives an approximation of the ℓ2-gain of the system. An illustrative example shows how well the algorithm performs compared to other methods approximating this quantity.

Place, publisher, year, edition, pages
Elsevier B.V., 2018
Keywords
excitation design, identification for control, Input, non-linear system identification, ℓ2-gain, Linear programming, Linear systems, Markov processes, Input-output data, Linear programs, Markov chain approaches, Stationary process, Target estimations, Nonlinear systems
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-247492 (URN)10.1016/j.ifacol.2018.09.095 (DOI)000446599200016 ()2-s2.0-85054461369 (Scopus ID)
Conference
18th IFAC Symposium on System Identification
Note

QC 20190418

Available from: 2019-04-18 Created: 2019-04-18 Last updated: 2020-03-04Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-0355-2663

Search in DiVA

Show all publications