Maintaining a relevant dataset for data-driven MPC using Willems' fundamental lemma extensions

Alexandre Faye-Bedrin (1, 2) , Stanislav Aranovskiy (1, 2) , Paul Chauchat (2, 1) , Romain Bourdais (1, 2)

1 CentraleSupélec [campus de Rennes]
2 IETR - Institut d'Électronique et des Technologies du numéRique

This work explores the recent formulation of nonlinear Data-driven Model Predictive Control in the case of dynamic references. Indeed, the state-of-the-art methods rely on Willems fundamental lemma, and freeze the used dataset at some point. While this ensures consistent behavior, i.e., excitation and accuracy, for a given setpoint, this will likely fail when the reference, and thus the operating point, changes. To this end, we propose refined heuristics for dataset management. First, a singular value-based method induces regular dataset updates but still guarantees a minimum excitation level. Then, a double-dataset formulation aims at decoupling accuracy and excitation issues and leverages the singular value-based one. These heuristics are validated in real-time experiments on a heat-blower system.

Keywords: predictive control; persistency of excitation; dataset management

Domains: Systems and Control [cs.SY]; Automatic

Alexandre Faye-Bedrin, Stanislav Aranovskiy, Paul Chauchat, Romain Bourdais. Maintaining a relevant dataset for data-driven MPC using Willems' fundamental lemma extensions. 2023. ⟨hal-04073063⟩