Fault detection and identification for quadrotor based on airframe vibration signals: A data-driven method

This paper proposes a new method to detect and identify rotor's fault of quadrotor by using airframe vibration signals. A three-level wavelet packet decomposition method is used to analyze vibration signals. Then, the standard deviations of wavelet packet coefficients construct feature vectors that are used as input signals to design a fault diagnostor based on Artificial Neural Network (ANN). Output signals of the fault diagnostor reflect rotor health status. Finally, the effectiveness and performance of the proposed method are validated by airframe vibration data collected from a hovering experiment of a quadrotor.