State estimation of nonlinear systems based on heterogeneous multiple models: Some recent theoretical results

Rodolfo Orjuela (1) , Benoît Marx (2) , José Ragot (2) , Didier Maquin (2)

1 MIPS - Modélisation, Intelligence, Processus et Système
2 CRAN - Centre de Recherche en Automatique de Nancy

State estimation of nonlinear systems plays an important role in several control engineering problems. Multiple model approach is an interesting way to cope with this relevant problem. Indeed, multiple models are recognized as a powerful modelling tool for nonlinear dynamic systems. In this framework, several realisations of multiple models can be considered for submodel interconnections. In contrast to the most popular results found in the multiple model literature, we consider here heterogeneous multiple models which allow to use submodels of different state space dimensions. Thanks to this fact, flexibility and generality can be introduced in the modelling stage. This paper provides survey of recent results in state estimation strategies based on heterogeneous multiple models. Different kinds of observers are investigated in order to improve the state estimation with respect to disturbance as well as unknown inputs. Theoretical results on the observers design and the state estimation error convergence are presented. Discussion and criticisms of the suggested approaches are also proposed and further research are pointed out.

Keywords: multiple models; state estimation; unknown input estimation; observers design

Domains: Automatic

Rodolfo Orjuela, Benoît Marx, José Ragot, Didier Maquin. State estimation of nonlinear systems based on heterogeneous multiple models: Some recent theoretical results. 7th Workshop on Advanced Control and Diagnosis, ACD'2009, Nov 2009, Zielona Gora, Poland. pp.CDROM. ⟨hal-00434301⟩