Faulty classification system for VTOL UAV acoustic signal using machine learning

Unmanned Aerial Vehicle (UAV) performance monitoring is essentialfor safety and efficientflight operation. The propeller, a key element in flying performance, is thefocus ofour research. As a vital part ofthe Vertical take-off and landing (VTOL) UAV flight mechanism, propeller failure could lead to hazardous incidents and increased maintenance costs. This paper introduces a user-friendly graphical user interface (GUI) development for the VTOL UAV propeller faulty classification system using the MATLAB Design App. The GUI, designed, enables the identification ofdifferent propeller conditions based on time-domain andfrequency-domain acousticalfeatures. Users can select their preferred features for faulty prediction using a specified supervised machine learning algorithm. Our study demonstrates that the GUI for propeller faulty classification can provide fast and highaccuracy real-timeflying performance insights, significantly improving the efficiency ofmonitoring work in UA V technology and aviation safety.