Unmanned Aerial Vehicle Sensing Data Anomaly Detection by Relevance Vector Machine

In recent years, Unmanned Aerial Vehicle (UAV) has been gaining more and more attention for military and civilian utilization. How to monitor its condition is a crucial problem. The accuracy of the sensing data is one of the basic requirements to achieve its correct condition. Thus, one kind of data anomaly detection approaches by Relevance Vector machine (RVM) is proposed in this article. By utilizing the visual UAV simulator FlgihtGear, the sensing data with anomalous data are generated. The effectiveness of the proposed approach for detecting the anomalous data contained in the sensing data is evaluated.