Human-Assisted Resilient Coordination for Automated Platoon With False Data Injection Attack

This paper investigates the resilient platooning problem of connected and automated vehicles (CAVs) under false-data injection (FDI) attacks. These attacks can alter the system state of any vehicle within a platoon at any time, presenting a significant challenge. To tackle this, we introduce a human-assisted resilient coordination (HARC) approach for CAVs, comprised of three core modules: a joint attack detection mechanism, a human-assisted coordination scheme, and a distributed optimization-based control module. The attack detection approach monitors vehicles at two distinct levels - local communication and global communication. Our human-assisted coordination strategy is designed not only to respond efficiently to attacks but also to ensure seamless coordination of the vehicle platoon, thereby preventing accidents and enhancing flexibility. By incorporating human input, we bolster the resilience and security of platoon coordination. Moreover, we present a distributed optimization-based control technique tailored specifically for constrained CAVs. A resilience constraint is designed and incorporated into the optimization problem, thereby empowering CAVs to identify adversarial communication channels via a localized communication detector. We provide experimental results to showcase the effectiveness of our proposed approach and to highlight its implications for resilient platooning under FDI attacks.