Fixed-time sliding mode based observer for nonlinear systems with unknown parameters and unknown inputs

Kokou A.A. Langueh (1) , Gang Zheng (2, 1) , Thierry Floquet (1)

1 CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189
2 DEFROST - Deformable Robots Simulation Team

It is of importance to design observers for multi-variable nonlinear systems with unknown parameters and partially driven by unknown inputs. Such a problem arises in systems subject to disturbances or with inaccessible inputs and in many applications such as parameter identification, fault detection and isolation or cryptography. In this paper, the problem of fixed-time observation for nonlinear dynamical systems with unknown parameter and inputs is studied. Conditions on full/partial state and parameters identification are provided by the the way of an observation algorithm based on differential geometry theory. Then, an uniform differentiator for estimating simultaneously the states and unknown parameters in fixed-time while avoiding observability singularities is designed. An example on topology identification of network systems is described to show the effectiveness of the proposed method.

Domaines: Automatique

Kokou A.A. Langueh, Gang Zheng, Thierry Floquet. Fixed-time sliding mode based observer for nonlinear systems with unknown parameters and unknown inputs. IET Control Theory and Applications, 2020, 14 (14), pp.1920 - 1927. ⟨10.1049/iet-cta.2019.1049⟩. ⟨hal-02637630⟩