Publication: Vibration-based fault detection in drones using mems vibration sensors and camera
Loading...
Date
2023-08
Authors
Dinagaran a/l Tamilselvam
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In recent years, utilization of drones or Unmanned Aerial Vehicles (UAVs) in various fields has risen drastically. With it, comes along the risk of safety or life
risking accidents when a drone faces faults or failures. In this project, research has been made on the widely experienced drone faults such as propulsion faults. The propulsion faults, namely motor failure and a hex-rotor drone has been adopted in this project as the case of study. In this project, vibration-based fault detection technique along with acoustic based fault detection technique has been implemented to detect the anomaly caused by motor failure in the drone. The fault detection using both the described technique has been extracted from the drone in both ground mode and thrust mode. Along with the sensors, this project also includes the vibration estimation using an event-based camera which translates the measurement of events from each pixel to frequency in Hertz. The event-based camera, namely Prophesee SilkyEvCam EvC3A is used detect the frequency of the vibrating arm of the motor. The detected vibration frequency from the camera is then compared with the data from the vibration sensor which is transformed from time domain to frequency domain beforehand. Increased
vibration was detected on the motor arm during the imbalanced propeller condition. Also, both vibration measurement of accelerometer and camera managed to detect different type of faults in the drone. By analysing the applicability of both the vibration-based detection technique this project aims to understand the likelihood of using event-based cameras to detect different faults in drones and compare vibration based measurement technique with data collected using acoustic-based measurement technique.