Project Abstract

In an era where autonomous vehicles are becoming increasingly prevalent, the intersection of vehicle technology and cybersecurity takes on critical importance. This research project uniquely positions itself by exploring the application of Reinforcement Learning (RL), a branch of Machine Learning, to enhance the security of automotive networks. By employing RL, the project aims to detect and defend against cyber threats more effectively than traditional static analysis or security-by-design methods.The methodology of the research consisted of understanding and identifying vulnerabilities in automotive microcontrollers and networks, exploring RL, designing an RL Framework, and evaluating the performance of RL agents in detecting and assessing security threats. The study concluded with listing the findings for further actions.The research has led to the discovery of the effectiveness of RL in identifying suspicious traffic and potential future cyber-attacks, providing a robust solution to the cybersecurity threats in the automotive industry. A significant finding was the ability of RL to learn during the process in an unknown environment, proving crucial in fulfilling the security and safety requirements.In conclusion, the research has contributed significantly to the field of automotive cybersecurity by presenting a systematic, dynamic approach to the protection of autonomous vehicles. The world now understands better the potential of RL as a defense mechanism against cyber threats, paving the way for safer and smarter transportation systems.

Keywords: , Cyber Security, Reinforcement Learning

 Conference Details

Session: Presentation Stream 2 at Presentation Slot 10

Location: GH043 at Tuesday 7th 13:30 – 17:00

Markers: Mukesh Tiwary, Muneeb Ahmad

Course: MSc Cyber Security, Masters PG

Future Plans: I’m looking for work