Convergence guarantees for adaptive model predictive control with kinky inference

Riccardo Zuliani, Raffaele Soloperto, John Lygeros
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:1058-1070, 2024.

Abstract

We analyze the convergence properties of a robust adaptive model predictive control algorithm used to control an unknown nonlinear system. We show that by employing a standard quadratic stabilizing cost function, and by recursively updating the nominal model through kinky inference, the resulting controller ensures convergence of the true system to the origin, despite the presence of model uncertainty. We illustrate our theoretical findings through a numerical simulation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-zuliani24a, title = {Convergence guarantees for adaptive model predictive control with kinky inference}, author = {Zuliani, Riccardo and Soloperto, Raffaele and Lygeros, John}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {1058--1070}, year = {2024}, editor = {Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}, volume = {242}, series = {Proceedings of Machine Learning Research}, month = {15--17 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v242/zuliani24a/zuliani24a.pdf}, url = {https://proceedings.mlr.press/v242/zuliani24a.html}, abstract = {We analyze the convergence properties of a robust adaptive model predictive control algorithm used to control an unknown nonlinear system. We show that by employing a standard quadratic stabilizing cost function, and by recursively updating the nominal model through kinky inference, the resulting controller ensures convergence of the true system to the origin, despite the presence of model uncertainty. We illustrate our theoretical findings through a numerical simulation.} }
Endnote
%0 Conference Paper %T Convergence guarantees for adaptive model predictive control with kinky inference %A Riccardo Zuliani %A Raffaele Soloperto %A John Lygeros %B Proceedings of the 6th Annual Learning for Dynamics & Control Conference %C Proceedings of Machine Learning Research %D 2024 %E Alessandro Abate %E Mark Cannon %E Kostas Margellos %E Antonis Papachristodoulou %F pmlr-v242-zuliani24a %I PMLR %P 1058--1070 %U https://proceedings.mlr.press/v242/zuliani24a.html %V 242 %X We analyze the convergence properties of a robust adaptive model predictive control algorithm used to control an unknown nonlinear system. We show that by employing a standard quadratic stabilizing cost function, and by recursively updating the nominal model through kinky inference, the resulting controller ensures convergence of the true system to the origin, despite the presence of model uncertainty. We illustrate our theoretical findings through a numerical simulation.
APA
Zuliani, R., Soloperto, R. & Lygeros, J.. (2024). Convergence guarantees for adaptive model predictive control with kinky inference. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:1058-1070 Available from https://proceedings.mlr.press/v242/zuliani24a.html.

Related Material