electro-hydrostatic-actuators

This paper presents an offline anomaly detection framework using an LSTM autoencoder for Electro-Hydrostatic Actuators (EHAs), achieving an average accuracy of 99.0% and high recall on sensor data, outperforming classical methods.