Unknown input observer based on discrete-time nonlinear descriptor model for understanding sitting control

Mathias Blandeau , Thierry-Marie Guerra (1) , Víctor Estrada Manzo , Philippe Pudlo , François Gabrielli

1 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201

The present paper studies sitting control for persons with complete thoracic spinal cord injury by means of automatic control. The methodology begins by modeling sitting control, after a discretization of the nonlinear descriptor model is computed, then a Takagi-Sugeno model is obtained for further manipulations. The approach uses unknown input observers to estimate states and internal variable of the system. The design conditions are expressed as linear matrix inequalities, whose solvability depends on convex optimization techniques. Simulations are conducted to show the effectiveness of the proposed observer.

Mots clés: Unknown input observer; Takagi-Sugeno models; Inverted pendulum; Spinal Cord Injury; Sitting movement; Nonlinear descriptor systems; Discrete-Time systems

Domaines: Automatique / Robotique

Mathias Blandeau, Thierry-Marie Guerra, Víctor Estrada Manzo, Philippe Pudlo, François Gabrielli. Unknown input observer based on discrete-time nonlinear descriptor model for understanding sitting control. 20th World Congress of the International Federation of Automatic Control, Jul 2017, Toulouse, France. pp. 820-825, ⟨10.1016/j.ifacol.2017.08.146⟩. ⟨hal-03413110⟩