Nowadays, UAVs (Unmanned aerial vehicles) have been used widely in various industries. UAVs are normally powered by a rotor installed with the propeller and often operated in complex environments which makes the UAV prone to failure. In order to reduce accidents and improve the reliability of UAVs, a number of methods have been proposed in this paper. Basically, these methods can be divided into model-based methods and data-driven methods. With the emergence of big data analysis, the data-driven method has attracted more and more attention. This method has demonstrated outstanding ability in the field of damage detection and identification. In this paper, a variety of damage detection and identification for UAVs based on data-driven methods are listed and compared. Both advantages and disadvantages of each method are discussed. Furthermore, the application of the data-driven method in other industries is also mentioned in the work. © 2022 The Aeronautical and Astronautical Society of the Republic of China. All rights reserved.