New fault tolerant control strategy for nonlinear systems with multiple model approach

This paper addresses a new methodology to construct a fault tolerant control (FTC) in order to compensate actuator faults in nonlinear systems. This approach is based on the representation of the nonlinear model with a multiple model under Takagi-Sugeno's form. The proposed control requires a simultaneous estimation of the system states and of the occurring actuator faults. The performance of the control depends on the quality of the estimations, indeed, it is important to estimate accurately and rapidly the states and the faults. This task is then performed with an Adaptive Fast State and Fault Observer (AFSFO). The stability conditions are established with Lyapunov theory and expressed in linear matrix inequality formulation to ease the design of the FTC. Furthermore, relaxed stability conditions are given with the use of the Polya's theorem.