Pose, Claudio Daniel (Universidad de Buenos Aires), Giribet, Juan Ignacio (Universidad de San Andrés), Torre, Gabriel (Universidad de San Andrés), Marzik, Guillermo (Universidad de San Andrés)

Neural Network-Based Propeller Damage Detection for Multirotors

Scheduled for presentation during the Regular Session "Fail-Safe Systems" (WeA1), Wednesday, June 7, 2023, 11:40−12:00, Room 118

2023 International Conference on Unmanned Aircraft Systems (ICUAS), June 6-9, 2023, Lazarski University, Warsaw, Poland

This information is tentative and subject to change. Compiled on May 11, 2025

Abstract

This work presents a method for detecting and identificating possible damages to propeller blades in multirotor vehicles, for a particular case study of a quadrotor. The detection method is based on a neural network, which takes as input the energy of several spectral bands of the inertial measurements and control variables, and outputs a measure of how damaged a propeller is. The ability of the network to correctly generalize from a limited dataset will be shown by training it using data gathered from an indoor, controlled environment, and testing it using data from outdoor flights.