Enhancing Autonomous Vehicle Security through Software-Defined Networking and Ai-Driven Threat Detection

Abstract

The increasing digital footprint of autonomous vehicles, in conjunction with the advent of technologies connected to artificial intelligence, has increased the potential for cyber vulnerabilities. This is a result of both factors. The three levels that comprise an autonomous driving system are the Sensor Layer, the Communication Layer, and the Control Layer. When seen from an attacker's perspective, these levels include the most crucial components. Regarding the sensor layer, some sensors constantly monitor the dynamics of the environment and the vehicle around it. On the other hand, these sensors are vulnerable to eavesdropping, jamming, and spoofing attacks. The communication layer combines both near-field and far-field communications to permit communication between more edge sensors in the area and faraway edge data centres. This connectivity is necessary to facilitate communication. The 'man-in-the-middle' assault and the Sybil attack can exploit this layer. Functionalities of an autonomous driving system, such as the automation of a vehicle's speed, braking, and steering, are made possible by the control layer, which is located at the very top of the hierarchy. Implementing artificial intelligence into software-defined vehicles paves the way for a wide range of options that could transform the landscape of automotive technology and mobility. Additionally, these prospects have the potential to revolutionize the industry.