A virtual actuator and sensor approach for fault tolerant control of a class of sampled-data control systems

Digital technologies have led to the widespread adoption of hybrid control systems, where sampled-data mechanisms introduce challenges in state estimation and control. While inter-sample predictor-based observers address these issues, they typically assume fault-free operation. In practice, sensor and actuator faults can degrade performance or cause instability. This paper proposes a fault-tolerant sampled-data control system that integrates virtual actuator and virtual sensor components within the inter-sample predictor framework. The virtual actuator follows a standard fault-hiding approach, while the virtual sensor employs a faulty inter-sample predictor that continuously reconstructs correct output samples, ensuring proper reinitialization of the nominal inter-sample predictor. This approach enhances robustness while preserving the benefits of inter-sample prediction. However, fault tolerance imposes a stricter constraint on the maximum allowable sampling time. Theoretical analysis and numerical simulations demonstrate the effectiveness of the proposed method.