AI-based anomaly identification techniques for vehicles communication protocol systems: Comprehensive investigation, research opportunities and challenges

The use of Controller Area Network in advanced automobiles as a communication technology is becoming more common. However, there is a lack of adequate privacy standards, such as data verification and cryptography. Therefore, the controller area network system is also susceptible to innumerable data breaches that can lead to serious consequences. To address this issue, multiple anomaly identification technologies were designed to identify these kinds of breaches. Nonetheless, the extraordinary standardisation characteristics of artificial intelligence enable anomaly detection techniques a viable preventative strategy against vehicle data security breach by the hackers. This study provides a comprehensive review of anomaly detection strategies facilitated by artificial intelligence and implemented between January 2018 and January 2024 It examines identification methodologies, threat varieties, characteristics, and standard information sets. Moreover, the paper also addresses the privacy concerns of AI designs, the prerequisites needed to design AI-based anomaly detection approaches in the controller area network channels, the constraints of currently available recommendations, and potential future study proposals. Additionally, to help researchers, and automobile manufacturers, in the rapidly growing field of protecting controller area network systems in modern car industries, the present research aims to shed light on the complicated landscape of anomaly detection within the domain of AI.