Robust Fault-tolerant Control via Linear Fractional Transformations

This paper presents a linear matrix inequality (LMI) approach to solve the robust fault-tolerant control problem against actuator failures. Using an equivalent transformation, fault effect factors can be put in Linear Fractional Transformation (LFT) form. While the fault effect factor, such as actuator fault, can be estimated accurately and swiftly, the fault-tolerant control (FTC) problem can be solved with a well-recognized design approach called gain-scheduling control theory and a fault-tolerant controller which provided with adaptive function can be designed for better performance of closed-loop system than passive robust fault-tolerant controller. The two-stage Kalman estimator which can estimate fault effect factors is utilized in this paper. The proposed design technique is finally evaluated in the light of a simulation example.