Networked Active Fault-Tolerant Predictive Control for Systems With Random Communication Constraints and Actuator/Sensor Faults

This brief presents an active fault-tolerant predictive control method for the output tracking of networked control systems (NCSs) subject to random communication constraints as well as actuator and sensor faults. The system state and actuator/sensor faults are simultaneously estimated via an observer and then used to perform a fault-tolerant predictive controller. Thus, both faults and the communication constraints in the backward and forward channels are actively compensated. A closed-loop stability condition is obtained, which also guarantees a zero steady-state tracking error. Moreover, the separation principle for designing the observer-based controller holds. Finally, the numerical simulation for a mass-spring-damper system is conducted to illustrate the effectiveness of the proposed method.