Robust fault-tolerant control of nonlinear switched stochastic systems based on functional observers

The fault-tolerant control problem for nonlinear switched stochastic systems affected by unknown inputs, sensor faults, and actuator faults is studied. Functional observers are adopted to reconstruct linear combinations of system states and faults, with the outcomes directly integrated into the controller synthesis process. The generalized inverse technique is utilized to decouple unknown inputs and disturbances. In contrast to conventional observer design strategies, the proposed scheme dispenses with the necessity of estimating all system states and faults, which remarkably decreases the number of estimated variables and thereby cuts down the computational cost of estimation. A fault-tolerant control framework based on functional observers is developed to guarantee the mean-square exponential stability of the closed-loop systems. Two simulation cases are conducted to demonstrate the feasibility and effectiveness of the proposed approach.