05/2012 - IEEE Transactions on Automatic Control

Deadline November 1, 2012

Special Issues on "Relaxation Methods in Identification and Estimation Problems"

The TC-CACSD in collaboration with the TC on System Identification and Adaptive Control chaired by Daniel Rivera, together with D. Regruto of Politecnico di Torino, are the Guest Editors of a Special Issue on "Relaxation methods in identification and estimation problems" for the IEEE Transactions on Automatic Control. The deadline for submission to the special issue is October 1st, 2012. The call for paper can be found at the IEEE TAC website and is reported here for ease of reading:

Special Issues on "Relaxation Methods in Identification and Estimation Problems" - Deadline November 1, 2012.

The subject of system identification has a long history, and it still remains one of the most active fields of research in the control community. In the literature at large, particular attention has been devoted in recent years to the convexification of estimation problems, and convexification has become one of the major topics in system identification. A number of different approaches have recently emerged in the optimization community to address the problem of approximating the global solution of some classes of nonconvex optimization problems. The common idea behind all these approaches is to construct specific convex relaxations, which are guaranteed to converge, under proper assumptions and conditions, to the global optima of the original nonconvex problem.

The aim of this special issue is to twofold: first, to highlight the fact that many challenging open problems in system identification and estimation can be reliably addressed via a convexification/relaxation approach; secondly, to show that the interplay between the optimization and the control communities can suggest new exciting research directions in identification and estimation problems. The topics relevant to this special issue include (but are not limited to) the following relaxation approaches to linear and nonlinear identification and estimation problems: LMI/SDP relaxations, L1-based sparsification approaches, probabilistic/randomized methods, rank/nuclear-norm minimization.

Guest Editors: Diego Regruto (Politecnico di Torino); Fabrizio Dabbene (CNR-IEIIT); Daniel E. Rivera (Arizona State University).