Secure Teleoperated Vehicles in Augmented Reality of Things: A Multichain and Digital Twin Approach
Sadia Jabeen Siddiqi, Mian Ahmad Jan, Anas Basalamah, Muhammad Tariq
Abstract
The integration of artificial intelligence-powered automotive electronics has revolutionized teleoperated vehicles through the adoption of Augmented Reality of Things (ARoT). This transformation is facilitated by Augmented Reality-based Heads-Up Displays (AR-HUDs) connected to vehicle sensors, which improve driver visibility with immersive overlays, taking in-vehicle human-computer interaction to new heights. These developments include Advanced Driver-Assistance Systems (ADAS) that aid drivers whether they are physically present in the vehicle or remotely controlling it. However, integrating ARoT and ADAS raises security and privacy concerns, especially regarding false data injection attacks. This paper analyzes security vulnerabilities in AR-HUDs, focusing on forged frames in shared AR-HUD video between teleoperated vehicles and their Digital Twins (DT). Critical vulnerabilities are identified through DT, necessitating effective mitigation for Metaverse access. In addition, the proposed solution is a decentralized data security framework based on multiple blockchains, overcoming latency and scalability issues with a Multichain scaffold. Furthermore, the ARoT heavily relies on a deep neural network for computer vision-based obstacle detection, with the results displayed on the AR-HUD. This helps secure the entire ARoT ecosystem, ensuring the integrity and safety of teleoperated vehicles.