Set-membership Switched Observers based on Interval Characterization of the Estimation Error

Sara Ifqir (1) , Vicenç Puig (2) , Dalil Ichalal (1) , Naïma Aït Oufroukh (1) , Saïd Mammar (1)

1 IBISC - Informatique, BioInformatique, Systèmes Complexes
2 UPC - Universitat Politècnica de Catalunya = Université polytechnique de Catalogne [Barcelona]

This paper presents a new set-membership estimation methodology for uncertain switched LPV discrete-time systems subject to unknown inputs, unmeasurable time-varying parameters and measurement noise. The proposed approach provides a guaranteed interval that is constructed as the sum of punctual state estimation and its corresponding estimation error limits. First, a punctual switched unknown input observer, robust against unknown inputs and time-varying uncertainties, is constructed. The proposed switched observer design is based on the solution of an optimization problem in terms of LMIs. Then, an outer-approximation of the enclosure set of state estimation error is computed using the admissible bounds of state and uncertainties. Application to vehicle state estimation is provided to show the design procedure and the flexibility of the proposed scheme. Comparison to real data demonstrates the accuracy and effectiveness of the obtained results.

Domains: Automatic

Sara Ifqir, Vicenç Puig, Dalil Ichalal, Naïma Aït Oufroukh, Saïd Mammar. Set-membership Switched Observers based on Interval Characterization of the Estimation Error. 21st IFAC World Congress (IFAC 2020), Jul 2020, Berlin, Germany. pp.14261--14266, ⟨10.1016/j.ifacol.2020.12.1165⟩. ⟨hal-02877939⟩