Sensor and Actuator Fault Detection and Isolation for Urban Air Mobility

In this paper, we propose a novel sensor and actuator fault detection and isolation (FDI) method for redundant urban air mobility systems to effectively utilize the system’s inherent redundancy for enhanced safety and fault tolerance. We introduce a hybrid redundancy scheme, integrating analytical and hardware redundancy, which enables the straightforward and reliable distinction between actuator and sensor faults utilizing a simple structure based on residual comparison. Furthermore, to accomplish simultaneous sensor measurement separation and fault diagnosis, a voting mechanism is incorporated, successfully preventing the propagation of fault effects to the flight controller. The proposed FDI methodology is designed to operate across triplex, duplex, and single-operational sensor configurations. To correctly isolate faults in the redundant structure, the concept of a virtual sensor is introduced in the duplex configuration. The performance of proposed method was verified through simulations, which showed its ability to successfully detect and isolate every fault under all considered configurations. The virtual sensor accurately determined sensor faults in the redundant architecture. Simultaneously, the voting mechanism ensured the accuracy of the control signals, effectively mitigating any adverse impact on the controller.