Strategizing for Cyber Security Enhancement Autonomous Intersection Management (AIM)

This research introduces a Reinforcement Learning (RL)-enabled framework to enhance the cybersecurity of Autonomous Intersection Management (AIM) systems. While AIM systems optimize traffic flow and reduce congestion, their reliance on interconnected networks makes them vulnerable to cyberattacks. Using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C), the framework dynamically detects, mitigates, and responds to cyber threats in a simulated AIM environment. Tested in a digital-twin intersection simulation, the RL agents effectively balanced traffic efficiency and security, handling threats like denial of service (DoS) attacks, data manipulation, and malware.Looking ahead, this framework sets the stage for scalable and adaptive cybersecurity solutions in intelligent transportation systems. It highlights RL’s potential to create resilient systems that evolve with emerging threats and emphasizes the value of explainable AI for transparency and trust. Future research can explore hybrid learning models, real-world datasets, and collaborations to establish robust security standards for AIM systems. This work underscores RL’s transformative potential to secure AIM systems and advance cybersecurity in transportation and critical infrastructure.