Research Projects

The "SafeCAV" project aims to address the growing security and scalability challenges in Connected Autonomous Vehicles (CAVs), which are rapidly becoming a key component of intelligent transportation systems. CAVs offer numerous benefits such as improved safety, precision in driving, and more efficient decision-making. However, their tightly coupled architecture, with high inter-dependency between perception modules, decision-making algorithms, communication, and control systems, exposes them to risks such as single points of failure. For example, incidents like Tesla's over-the-air (OTA) software updates, which inadvertently introduced bugs affecting vehicle performance, highlight the critical need for more robust security measures. Furthermore, the rapid rise of Vehicle-to-Everything (V2X) improves safety and operational efficiency, it also introduces significant cybersecurity risks, particularly regarding remote vehicle hijacking in DSRC under the IEEE 1609.2 protocol. The dynamic, heterogeneous, and distributed nature of the V2X environment makes traditional static security models insufficient. The continuous movement of vehicles across network boundaries and the diverse communication protocols used further complicate the protection of these systems. To tackle these issues, SafeCAV proposes a shift to the Zero Trust Architecture (ZTA) principle, a security model based on the principle of "Never Trust, Always Verify." SafeCAV will obey the ZTA principle with container orchestration to secure CAV communications and prevent unauthorized access. A key feature of the project is the development of a lightweight framework for deploying microservices on CAVs, utilizing a decentralized Software-Defined Vehicular Network (SDVN) architecture for monitoring and localized command issuance through programmable control by avoiding single points of failure. Additionally, SafeCAV will implement a two-level authentication protocol for secure microservice execution: Level-1 ensures container integrity through origin verification and manufacturer certification, and Level-2 deploys a lightweight in-container authentication mechanism to counter unauthorized access and remote hijacking ensuing isolated execution. A trusted CI/CD pipeline using Jenkins and Docker Swarm will be designed to enable the dynamic deployment of these microservices, automating the process of building, deploying, and managing containers. To optimize deployment, the pipeline will employ tools like DockerSlim, BuildKit, and Trivy to reduce image sizes, streamline builds, and ensure runtime efficiency. Finally, SafeCAV will validate these security measures through a plugin-based hardware ADAS prototype with TRL of 4–5 to validate the anti-remote hijacking mechanisms using a trusted container CI/CD pipeline to adhere to the microservice framework. This prototype will demonstrate the effectiveness of the microservice deployment framework and anti-hijacking protocols in a real-world setting.