Experiment design for batch-to-batch learning control

M. Forgione, X. Bombois, P.M.J. Hof, Van den

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

An Experiment Design framework for dynamical systems which execute multiple batches is presented in this paper. After each batch, a model of the system dynamics is refined using the measured data. This model is used to synthesize the controller that will be applied in the next batch. Excitation signals may be injected into the system during each batch. From one hand, perturbing the system worsens the control performance during the current batch. On the other hand, the more informative data set will lead to a better identified model for the following batches. The role of Experiment Design is to choose the proper excitation signals in order to optimize a certain performance criterion defined on the set of batches that is scheduled. A total cost is defined in terms of the excitation and the application cost altogether. The excitation signals are designed by minimizing the total cost in a worst case sense. The Experiment Design is formulated as a Convex Optimization problem which can be solved efficiently using standard algorithms. The applicability of the method is demonstrated in a simulation study.
Original languageEnglish
Title of host publicationProceedings of the 2013 American Control Conference (ACC 2013), 17-19 July 2013, Washington D.C.
Pages3198-3923
Publication statusPublished - 2013
Event2013 American Control Conference (ACC 2013), June 17-19, 2013, Washington, DC, USA - Renaissance Washington, DC Downtown Hotel, Washington, DC, United States
Duration: 17 Jun 201319 Jun 2013
http://acc2013.a2c2.org/

Conference

Conference2013 American Control Conference (ACC 2013), June 17-19, 2013, Washington, DC, USA
Abbreviated titleACC 2013
Country/TerritoryUnited States
CityWashington, DC
Period17/06/1319/06/13
Internet address

Fingerprint

Dive into the research topics of 'Experiment design for batch-to-batch learning control'. Together they form a unique fingerprint.

Cite this