Multiaxial vibration data for blade fault diagnosis in multirotor unmanned aerial vehicles.

This dataset presents multiaxial vibration signals collected from a multirotor unmanned aerial vehicle (UAV) operating in hover mode for the purpose of blade fault diagnosis. Vibration measurements were recorded at the geometric center of the UAV, where the centerlines of the four rotor arms intersect, using a triaxial accelerometer. The dataset captures variations across the X, Y, and Z axes under different blade fault conditions, including healthy, minor imbalance, severe imbalance, and screw loosening scenarios. Each flight scenario was repeated under controlled conditions to ensure consistency and high-quality labeling. The resulting soft-labeled dataset includes time-domain signals from numerous test flights and has been used in multiple prior studies involving classical and deep learning-based fault classification techniques. This curated data collection provides a valuable resource for researchers in UAV health monitoring, vibration analysis, and machine learning-based fault diagnosis. The dataset is particularly useful for the development and benchmarking of signal processing pipelines and classification models aimed at identifying blade-level faults in multirotor UAV systems.