Generative AI Models and LLMs for Transportation Cybersecurity
Project Description
Modern transportation systems, ranging from intelligent traffic management to autonomous vehicles, are increasingly vulnerable to cyber threats due to their reliance on interconnected digital infrastructures. Traditional rule-based cybersecurity measures often fail to detect sophisticated and evolving attacks. This research proposes leveraging capabilities of Generative AI models and Large Language Models (LLMs) to enhance the detection, prediction, and mitigation of cyber risks in autonomous transportation networks.
The objectives are threefold: 1) Develop Gen AI and LLM-driven methods to identify anomalies in system logs, communication protocols, and sensor data, 2) Explore LLMs for real-time threat intelligence, including detecting phishing and malicious command injection, and 3) Design adaptive defense strategies where LLMs assist in simulating adversarial scenarios and recommending countermeasures.
We will integrate domain-specific fine-tuning of LLMs with transportation cybersecurity datasets. Experiments will involve simulating cyberattacks on vehicular networks and intelligent transportation systems, followed by evaluation of LLM-based detection and response. This research will provide scalable, AI-enhanced cybersecurity solutions for critical transportation infrastructures, improving resilience against emerging cyber threats while ensuring safety and reliability in next-generation mobility systems.
Start Date
January or May 2026
Postdoc Qualifications
Background in AI methods like RL, Gen AI
Background in Cybersecurity
PhD in ECE, CS or Transportation Systems
Interest in Autonomous Transportation
Co-advisors
Satish Ukkusuri (CE and CS)
Lingxi Li (ECE)
Bibliography
1. D4: Dynamic Data-Driven Discovery
C Hernandez¹, DO Barbosa, Z Lei, L Burbano… - Dynamic Data Driven Applications Systems: 5th …, 2025
2. Cybersecurity for next-generation road transportation: A review
S Ukkusuri, OF Hamim, Z Lei, E Ka, MS Salek… - Journal on Autonomous Transportation Systems, 2025
3. https://arxiv.org/html/2404.11338v1
4. https://arxiv.org/html/2403.08701v2
5. https://scholar.google.com/citations?view_op=view_citation&hl=en&user=9gmoT80AAAAJ&sortby=pubdate&citation_for_view=9gmoT80AAAAJ:YTuZlYwrTOUC