Stability and robustness of nonlinear predictive control without stabilizing terminal constraints
Résumé
We developed a stability and guaranteed performance analysis method for MPC schemes with which we can compute optimization horizon bounds N under controllability assumptions. The approach can be coupled with robust MPC variants the method can be extended to analyzing varying control horizons. Main conclusion: larger and varying control horizons can be used without losing (nominal) stability and performance. However, longer control horizons may reduce robustness. Tasks in SADCO project: improve robustness using sensitivity techniques and integrated stability and robustness analysis.
Domaines
Optimisation et contrôle [math.OC]
Origine : Fichiers produits par l'(les) auteur(s)