Actuator and Sensor Fault Detection and Diagnosis for Unmanned Quadrotor Helicopters

Abstract

In this paper, the problem of actuator and sensor fault detection and diagnosis (FDD) for unmanned quadrotor helicopters (UQHs) is addressed by utilizing the interacting multiple model (IMM) approach. The mathematical model of UQHs, including actuator and sensor faults, is firstly introduced. Actuator faults are modelled as the loss of control effectiveness in motors and sensor faults are assumed to be bias fault. In order to reduce the intensive computational load due to the large fault model set, a state augmentation (SA) strategy is applied. This strategy can not only reduce the number of elemental filter, but also identify the magnitude of faults. Through combining the IMM algorithm and the SA strategy, the proposed FDD scheme can effectively detect and diagnose multiple actuator and sensor faults. Finally, the effectiveness of the FDD scheme is validated by two simulation scenarios, and the obtained results prove that the proposed scheme is an effective tool to deal with actuator and sensor faults.