A new paper has been published by the SDA-Lab in the prestigious journal ‘Engineering Applications of Artificial Intelligence’ (Impact Factor: 8, Q1). Titled ‘An Unsupervised Anomaly Detection Framework for Onboard Monitoring of Railway Track Geometrical Defects Using One-Class Support Vector Machine,’ this paper introduces an innovative framework for detecting geometrical defects in railway tracks through onboard monitoring, employing a one-class support vector machine. The authors of this paper are Dr. Ramin Ghiasi, Dr. Muhammad Arslan Khan, Dr. Danilo Sorrentino, Cassandre Diaine, and Dr. Abdollah Malekjafarian.
This collaborative work with SNCF Réseau involved implementing the proposed framework on real data collected from the IRIS320 high-speed train in France.
The full paper is available open access from this link:
https://www.sciencedirect.com/science/article/pii/S0952197624003257