← Back to list of papers of the 2024 EuroGNC conference |
CEAS EuroGNC 2024 |
||||||
A Scenario-based Approach to Robust Control Design |
||||||
|
||||||
Abstract: This paper proposes control tuning strategies for an ensemble of scenarios. Each of these scenarios might correspond to a different open-loop plant and/or an operating condition. Overfitting is prevented by ensuring that the closed-loop requirements are satisfied when the values prescribing such scenarios are perturbed. To this end we first model the perturbed scenarios as sample sets of finite size. Relaxed chance-constrained optimization is then used to seek controllers with varying degrees of robustness. For instance, we can deliberately eliminate a given number of scenarios from the dataset in order to obtain a riskier controller with a better performance, or we might seek a conservative controller that satisfies the closed-loop requirements with an acceptably high probability for as many perturbed scenarios as possible. The scenarios for which the requirements are not met, by either physics-based limitations or choice, are optimally chosen while the controller gains are searched for. The design of a feedback control system having a non-collocated sensor-actuator pair with time domain requirements is used for illustration | ||||||
Keywords: Data-driven control; chance-constrained optimization; uncertainty; robustness; outlier elimination | ||||||
Luis G. Crespo, Tanner Slagel, Sean P. Kenny: A Scenario-based Approach to Robust Control Design. Proceedings of the 2024 CEAS EuroGNC conference. Bristol, UK. June 2024. CEAS-GNC-2024-057. |
BibTeX entry: @Incollection{CEAS-GNC-2024-057, author = {Crespo, Luis G. and Slagel, Tanner and Kenny, Sean P.}, title = {A Scenario-based Approach to Robust Control Design}, booktitle = {Proceedings of the 2024 {CEAS EuroGNC} conference}, address = {Bristol, UK}, month = jun, year = {2024}, note = {CEAS-GNC-2024-057} } |