Exploration and Detection of Cyber-Attacks on MPC-Based Vehicular Platoons

作者：Tashfique Hasnine Choudhury, Chengwei Duan, Hanyu Zhang, Lili Du, Sandip Ray · 年份：2024 · DOI：10.1109/itsc58415.2024.10919989 · 被引用次数：1 · 研究领域：Vehicular Ad Hoc Networks (VANETs)、Advanced Malware Detection Techniques、Cryptographic Implementations and Security

We develop an extensible platform for exploring cyber-attacks on connected and autonomous vehicle (CAV) platoon controllers. A unique feature of our platform is the ability to address a large spectrum of cyber threats, including new attacks designed after deployment. This platform facilitates systematic exploration of cyber-attacks, helping to uncover real-time vulnerabilities across different designs of platoon controllers. Additionally, we introduce a novel, dual-stage detection mechanism that employs a time-series machine learning (ML) model for the real-time detection of attacks on the perception channels of a Model Predictive Control (MPC)-based CAV platoon. We present extensive experimental results to showcase the effectiveness and feasibility of our approach.