RflyMAD: A dataset for multicopter fault detection and health assessment

Abstract

Listen

Translate article

This paper presents an open-source dataset, RflyMAD, a Multicopter Abnormal Dataset developed by the Reliable Flight Control (Rfly) Group aiming to promote the development of research fields such as Fault Detection and Isolation (FDI) or Health Assessment (HA). The full 114 GB dataset includes 11 types of faults under 6 flight statuses which are adapted from the ADS-33 file to cover more cases where the multicopters have different levels of mobility when faults occur. In the total of 5629 flight cases, the fault time is up to 3283 min, and there are 2566 cases for software-in-the-loop (SIL) simulation, 2566 cases for hardware-in-the-loop (HIL) simulation, and 497 cases for real flight. As it contains simulation data based on RflySim and real flight data, it is possible to improve the quantity while increasing the quality of the data. In each case, there are ULog, Telemetry log, Flight information, and processed files for researchers to use and review. The RflyMAD dataset could be used as a benchmark for fault diagnosis methods and the support relationship between simulation data and real flight is verified by transfer learning methods. In the future, more methods will be presented as a baseline and RflyMAD will be updated with more data and types. In addition, the dataset and associated toolkit are available at https://rfly-openha.github.io/documents/4_resources/dataset.html .