A fault injection platform for multirotor UAV PHM

Abstract

Unmanned Aerial Vehicle (UAV) is widely used in daily life, commercial and military applications, because it is flexible and cheap. But the news of UAV accidents calls people to focus on Prognostics and Health Management (PHM) for UAV, which can predict and detect faults during the UAV ’s flight through the telemetry data. PHM can also help UAV adjust itself to avoid crashes. However, the existing PHM models on UAV have some common problems. The most outstanding one is that they are all suffering from lacking fault data. The amount of fault data is difficult to support PHM modeling. Therefore, establishing a fault injection platform will help in generating fault data and improving the performance of PHM methods. In this work, we propose a platform of fault injection based on ArduPilot. It can simulate different kinds of sensor and actuator faults which may happen on UAV during flight. Through the ground station, we can obtain the telemetry data with faults which can set up a database for PHM modeling. The experiment results show that this method has a good performance for fault injection.