Simulation-to-reality UAV Fault Diagnosis in windy environments

TL;DR

This paper introduces an uncertainty-based fault classifier for UAVs that effectively transfers from simulation to real windy environments, achieving high accuracy with limited data.

Contribution

It proposes a novel uncertainty-based fault classifier using ensemble deep CNNs to improve sim-to-real UAV fault diagnosis in windy conditions.

Findings

Achieves 100% fault diagnosis accuracy in windy outdoor scenarios.

Reduces data usage to 33.6% compared to traditional methods.

Employs ensemble and uncertainty filtering to enhance robustness.

Abstract

Monitoring propeller failures is vital to maintain the safe and reliable operation of quadrotor UAVs. The simulation-to-reality UAV fault diagnosis technique offer a secure and economical approach to identify faults in propellers. However, classifiers trained with simulated data perform poorly in real flights due to the wind disturbance in outdoor scenarios. In this work, we propose an uncertainty-based fault classifier (UFC) to address the challenge of sim-to-real UAV fault diagnosis in windy scenarios. It uses the ensemble of difference-based deep convolutional neural networks (EDDCNN) to reduce model variance and bias. Moreover, it employs an uncertainty-based decision framework to filter out uncertain predictions. Experimental results demonstrate that the UFC can achieve 100% fault-diagnosis accuracy with a data usage rate of 33.6% in the windy outdoor scenario.