Directory Intranet

COSIP - Control Systems and Identification Problems

Project Coordinator : Christophe PRIEUR

Project realized thanks to the support of : Autre organisme (Programme régional de coopération scientifique MATH-AmSUD)

Start date : 2014/01/01

Duration : 24 mounths

The main goal of this programm is to create and to consolidate scientific connections within South America and with France. In this perspective, this project have the following scientific goals:
- Study of the control properties of some coupled systems describing the evolution of plates (Mindlin-Timoshenko systems), the interactions between short-wave and long-wave (Schrodinger-KdV systems), and the propagation of surface and internal waves (Boussinesq systems).
- Characterization, for control systems where communication takes place over nontransparent communication links, of the interplay between optimal performance and concurrent communication constraints that includes delays.
- Study of the problem of identification of medium properties by using external measurements in the framework of electro-seismic prospection, semiconductor material and transmission phenomena for beams or plates.

The focus of this project will be in both control and identification problems.
Concerning control theory, we plan to work in two main research lines: controllability of coupled systems of partial differential equations (PDE) and stability of networked control systems under concurrent communication constraints. The interest in the study of control properties of coupled systems is rather recent but very active. There are a number of control results for single PDE but new phenomena appear when dealing with coupled systems. We intend to address this issue for both dispersive and hyperbolic coupled systems. On the other hand, for control systems where communication takes place over non-transparent communication links (networked control systems) we aim at addressing optimal control problems subject to concurrent communication constraints that include delays and/or are subject to asynchronous sampling.

Regarding identification problems, we address the recovering of heterogeneous medium properties (for instance conductivity, elasticity, charge density or transmission velocity) by means of indirect measurements. We look for stability properties and reconstruction algorithms. Very useful tools are the Carleman estimates, which in fact are useful as well to deal with control properties of PDE. We plan to develop new Carleman estimates at both the continuous level (including limiting Carleman weights) and the discrete one in order to address some identification problems.


This project involve this partners :


Eduardo Cerpa, Universidad Técnica Federico Santa María
Matías Courdurier, Pontificia Universidad Católica de Chile
Galina García, Universidad de Santiago
Alberto Mercado, Universidad Técnica Federico Santa María
Axel Osses, Universidad de Chile
Eduardo Silva, Universidad Técnica Federico Santa María

Fagner Araruna, Universidade Federal da Paraiba
Ademir Pazoto, Universidade Federal do Rio de Janeiro
Adán Corcho, Universidade Federal do Rio de Janeiro

Christophe Prieur, GIPSA-lab
Lucie Baudouin, Laboratoire d'Analyse et d'Architecture des Systèmes
Eric Bonnetier, Laboratoire Jean Kuntzmann, Université Joseph Fourier
Emmanuelle Crépeau, Université de Versailles St-Quentin
Sylvain Ervedoza, CNRS and Université Paul Sabatier
Frédéric Gouaisbaut, Laboratoire d'Analyse et d'Architecture des Systèmes
Alexandre Seuret, Laboratoire d'Analyse et d'Architecture des Systèmes
Sergio Guerrero, Université Pierre et Marie Curie


Projet financé dans le cadre du programme régional de coopération France-Amérique du Sud.

GIPSA-lab, 11 rue des Mathématiques, Grenoble Campus BP46, F-38402 SAINT MARTIN D'HERES CEDEX - 33 (0)4 76 82 71 31