Outlier robust system identification: A Bayesian kernel-based approach
2014 (English)In: IFAC Proceedings Volumes (IFAC-PapersOnline), IFAC Papers Online, 2014, 1073-1078 p.Conference paper (Refereed)
In this paper, we propose an outlier-robust regularized kernel-based method for linear system identification. The unknown impulse response is modeled 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. To build robustness to outliers, we model the measurement noise as realizations of independent Laplacian random variables. The identification problem is cast in a Bayesian framework, and solved by a new Markov Chain Monte Carlo (MCMC) scheme. In particular, exploiting the representation of the Laplacian random variables as scale mixtures of Gaussians, we 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.
Place, publisher, year, edition, pages
IFAC Papers Online, 2014. 1073-1078 p.
IdentifiersURN: urn:nbn:se:kth:diva-175122DOI: 10.3182/20140824-6-ZA-1003.01587ScopusID: 2-s2.0-84929727146ISBN: 9783902823625OAI: oai:DiVA.org:kth-175122DiVA: diva2:876186
19th IFAC World Congress on International Federation of Automatic Control, IFAC 2014, 24 August 2014 through 29 August 2014
QC 201512022015-12-022015-10-092015-12-02Bibliographically approved