Change search
Link to record
Permanent link

Direct link
BETA
Bottegal, Giulio
Publications (4 of 4) Show all publications
Risuleo, R. S., Bottegal, G. & Hjalmarsson, H. (2019). Modeling and identification of uncertain-input systems. Automatica, 105, 130-141
Open this publication in new window or tab >>Modeling and identification of uncertain-input systems
2019 (English)In: Automatica, ISSN 0005-1098, E-ISSN 1873-2836, Vol. 105, p. 130-141Article in journal (Refereed) Published
Abstract [en]

We present a new class of models, called uncertain-input models, that allows us to treat system-identification problems in which a linear system is subject to a partially unknown input signal. To encode prior information about the input or the linear system, we use Gaussian-process models. We estimate the model from data using the empirical Bayes approach: the hyperparameters that characterize the Gaussian-process models are estimated from the marginal likelihood of the data. We propose an iterative algorithm to find the hyperparameters that relies on the EM method and results in decoupled update steps. Because in the uncertain-input setting neither the marginal likelihood nor the posterior distribution of the unknowns is tractable, we develop an approximation approach based on variational Bayes. As part of the contribution of the paper, we show that this model structure encompasses many classical problems in system identification such as Hammerstein models, blind system identification, and cascaded linear systems. This connection allows us to build a systematic procedure that applies effectively to all the aforementioned problems, as shown in the numerical simulations presented in the paper.

Place, publisher, year, edition, pages
Elsevier, 2019
Keywords
Estimation algorithms, Gaussian processes, Nonlinear models, Nonparametric identification, System identification, Gaussian noise (electronic), Identification (control systems), Iterative methods, Linear systems, Religious buildings, Blind system identification, Empirical Bayes approach, Estimation algorithm, Non-linear model, Non-parametric identification, Posterior distributions, System identification problems, Gaussian distribution
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-252499 (URN)10.1016/j.automatica.2019.03.014 (DOI)000476963500013 ()2-s2.0-85063903295 (Scopus ID)
Note

QC 20190711

Available from: 2019-07-11 Created: 2019-07-11 Last updated: 2019-08-12Bibliographically approved
Zamani, M., Bottegal, G. & Anderson, B. D. O. (2016). On the Zero-Freeness of Tall Multirate Linear Systems. IEEE Transactions on Automatic Control, 61(11), 3606-3611
Open this publication in new window or tab >>On the Zero-Freeness of Tall Multirate Linear Systems
2016 (English)In: IEEE Transactions on Automatic Control, ISSN 0018-9286, E-ISSN 1558-2523, Vol. 61, no 11, p. 3606-3611Article in journal (Refereed) Published
Abstract [en]

In this technical note, tall discrete-time linear systems with multirate outputs are studied. In particular, we focus on their zeros. In systems and control literature zeros of multirate systems are defined as those of their corresponding time-invariant systems obtained through blocking of the original multirate systems. We assume that blocked systems are tall, i.e., have more outputs than inputs. It is demonstrated that, for generic choice of the parameter matrices, linear systems with multirate outputs generically have no finite nonzero zeros. However, they may have zeros at the origin or at infinity depending on the choice of blocking delay and the input, state and output dimensions.

Place, publisher, year, edition, pages
IEEE, 2016
Keywords
MIMO, multirate, zeros, zero-freeness
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-199548 (URN)10.1109/TAC.2016.2527924 (DOI)000389892000036 ()2-s2.0-84994850945 (Scopus ID)
Note

QC 20170116

Available from: 2017-01-16 Created: 2017-01-09 Last updated: 2017-11-29Bibliographically approved
Bottegal, G., Pillonetto, G. & Hjalmarsson, H. (2015). Bayesian kernel-based system identification with quantized output data. IFAC-PapersOnLine, 48(28), 455-460
Open this publication in new window or tab >>Bayesian kernel-based system identification with quantized output data
2015 (English)In: IFAC-PapersOnLine, ISSN 2405-8963, Vol. 48, no 28, p. 455-460Article in journal (Refereed) Published
Abstract [en]

In this paper we introduce a novel method for linear system identification with quantized output data. We model the impulse response as a zero-mean Gaussian process whose covariance (kernel) is given by the recently proposed stable spline kernel, which encodes information on regularity and exponential stability. This serves as a starting point to cast our system identification problem into a Bayesian framework. We employ Markov Chain Monte Carlo (MCMC) methods to provide an estimate of the system. In particular, we show how to design a Gibbs sampler which quickly converges to the target distribution. Numerical simulations show a substantial improvement in the accuracy of the estimates over state-of-the-art kernel-based methods when employed in identification of systems with quantized data.

Place, publisher, year, edition, pages
Elsevier, 2015
Keywords
Impulse response, Linear systems, Markov processes, Numerical methods, Religious buildings, Bayesian frameworks, Gibbs samplers, Identification of systems, Kernel based methods, Markov chain Monte Carlo method, State of the art, System identification problems, Zero mean Gaussian process, Monte Carlo methods
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-195439 (URN)10.1016/j.ifacol.2015.12.170 (DOI)2-s2.0-84988452886 (Scopus ID)
Note

QC 20161114

Available from: 2016-11-14 Created: 2016-11-03 Last updated: 2016-11-14Bibliographically approved
Everitt, N., Bottegal, G., Rojas, C. R. & Hjalmarsson, H. (2015). On the Effect of Noise Correlation in Parameter Identification of SIMO Systems. IFAC-PapersOnLine, 48(28), 326-331
Open this publication in new window or tab >>On the Effect of Noise Correlation in Parameter Identification of SIMO Systems
2015 (English)In: IFAC-PapersOnLine, ISSN 2405-8963, Vol. 48, no 28, p. 326-331Article in journal (Refereed) Published
Abstract [en]

The accuracy of identified linear time-invariant single-input multi-output (SIMO) models can be improved when the disturbances affecting the output measurements are spatially correlated. Given a linear parametrization of the modules composing the SIMO structure, we show that the correlation structure of the noise sources and the model structure of the othe modules determine the variance of a parameter estimate. In particular we show that increasing the model order only increases the variance of other modules up to a point. We precisely characterize the variance error of the parameter estimates for finite model orders. We quantify the effect of noise correlation structure, model structure and signal spectra.

Place, publisher, year, edition, pages
Elsevier, 2015
Keywords
Identification (control systems), Correlation structure, Effect of noise, Linear parametrization, Linear time invariant, Noise source, Parameter estimate, Single input multi outputs, Variance error, Parameter estimation
National Category
Control Engineering
Identifiers
urn:nbn:se:kth:diva-195438 (URN)10.1016/j.ifacol.2015.12.148 (DOI)2-s2.0-84988474760 (Scopus ID)
Note

QC 20161118

Available from: 2016-11-18 Created: 2016-11-03 Last updated: 2016-11-18Bibliographically approved
Organisations

Search in DiVA

Show all publications