An approximate characterisation of the set of feasible trajectories for constrained flat systems - CentraleSupélec Accéder directement au contenu
Article Dans Une Revue Automatica Année : 2022

An approximate characterisation of the set of feasible trajectories for constrained flat systems

Luca Greco
Hugues Mounier
Maria Bekcheva
  • Fonction : Auteur

Résumé

We study differentially flat systems, subject to constraints on the input, the state and their derivatives. We are interested in providing an approximate characterisation of the set of trajectories satisfying both the equations of the system and the constraints. The flatness property and the choice of a polynomial parametrisation in terms of Bézier functions enable a finite dimension formulation of the problem in the parameter space of the Bézier control points. Such formulation provides a unified method to express a large variety of relevant constraint types such as, for instance, physical obstacles, actuator limitations, tubes around nominal trajectories, performance requirements and energy restrictions. We present a simple and effective inner approximation of the aforementioned set of trajectories obtained by simple algebraic manipulations of the original constraints formulation. The symbolic nature of the proposed characterisation allows a certain degree of adaptation to changes in the constraints without the need for any re-computation. The entire method (formulation an characterisation) is illustrated via a classical benchmark mechanical system.
Fichier principal
Vignette du fichier
FlatConstr.pdf (6.94 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03692611 , version 1 (09-06-2022)

Identifiants

Citer

Luca Greco, Hugues Mounier, Maria Bekcheva. An approximate characterisation of the set of feasible trajectories for constrained flat systems. Automatica, 2022, 144, pp.110484. ⟨10.1016/j.automatica.2022.110484⟩. ⟨hal-03692611⟩
44 Consultations
115 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More