Vibration-Based Fault Detection in Drone Using Artificial Intelligence

Abstract

Recent years have seen a huge increase in the study of drones. There is a lot of published articles regarding drone, focusing on control optimization, fault detection, safety mechanisms, etc. In fault detection, most studies focused on the effects of faulty propellers and rotors, and there is very limited academic research on drone arms. In this paper, a fault detection based on the vibration of the multirotor arms using artificial intelligence (AI) is proposed. There are some cases in which, due to accident, the arm of the multirotor crack or loosen. This is normally unnoticeable without disassembly, and if not taken care of, it would have likely resulted in a sudden loss of flight stability, which will lead to a crash. Different types of AI methods are incorporated in this study, namely, fuzzy logic, neuro-fuzzy, and neural network (NN). Their results are compared to determine the best method in predicting the safety of the multirotor. Fuzzy logic and neuro-fuzzy methods provided acceptable decision-making, but the performance of the neuro-fuzzy approach depend on the dataset used because overfit model might give incorrect decision-making. This also applies to the NN technique. Because the vibration data are collected in the laboratory environment without consideration of wind effect, this framework is more suitable for early prediction before flying the multirotor in the outdoor environment.