Machine learning detects rotor faults and damage severity in VTOL aircraft
A Machine Learning Framework for Fault Detection, Isolation, and Severity Prediction of Autonomous VTOL Aircraft
Machine Learning
Summary
Small problems in the rotors of autonomous VTOL aircraft can cause unsafe flying conditions. Detecting these faults early is hard because of noisy sensors and complex flight physics. The authors created a machine learning method that looks at flight data to find which rotor is damaged and how badly. They trained it with simulated and real flight tests, reaching very high accuracy. This helps monitor the health of drones in real time to keep them safe.
What this means in practice
- •For drone operators: Continuously monitor rotor health during autonomous VTOL flight to detect and classify damage before failures occur.
- •For robotics engineers: Incorporate learned CNN models into control systems to improve fault-tolerant flight and fault diagnosis.
Authors
Ripon C. Sarker, Pedram H. Dabaghian, Raman Goyal, Atanu Halder
Abstract
Fault detection in autonomous VTOL aircraft is critical because even minor component degradations can rapidly destabilize multirotor vehicles operating in complex, safety-critical environments, motivating robust fault detection and estimation strategies capable of identifying early signs of rotor damage; however, real-flight fault detection remains challenging due to sensor noise, environmental disturbances, and the nonlinear aerodynamics of multirotor platforms. This study proposes a comprehensive machine-learning framework for rotor fault detection, isolation, and severity prediction using real flight data. A convolutional neural network (CNN) architecture is developed to learn spatio-temporal patterns from multivariate flight dynamics, enabling direct inference of both the faulted rotor and its damage level. The framework is first validated using simulated data generated by a data-generative model, and experimental validation is then performed on a hexacopter by introducing controlled blade-tip breakage. The trained model achieves rotor-wise fault classification accuracies above 99% and severity estimation accuracy of 96% within a 1% tolerance in experimental data, demonstrating strong generalization and supporting real-time health monitoring for autonomous VTOL systems.