Machine Learning for Anomaly Detection in Automotive Cyber-Physical Systems
Published in Springer Nature, 2023
Recommended citation: V. K. Kukkala, S. V. Thiruloga, and S. Pasricha, "Machine Learning for Anomaly Detection in Automotive Cyber-Physical Systems," in Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing, Springer Nature, 2023.
Abstract
Modern-day cars rely on powerful embedded systems known as electronic control units (ECUs) to control different components in the vehicle. The increasing efforts to make vehicles fully autonomous resulted in relying on information from various external sources, which made the ECUs in the vehicles highly vulnerable to various cyber threats. In this chapter, we introduce a novel deep learning-based anomaly detection framework called INDRA that utilizes a gated recurrent unit (GRU)-based recurrent autoencoder network to detect cyber-attacks in automotive cyber-physical systems. Our proposed INDRA framework is evaluated under different attacks and compared against various state-of-the-art anomaly detection works using a commercially available vehicular network dataset.