Explore the critical security implications of Autonomous Driving Domain Controllers in modern smart cars during this 33-minute Black Hat conference talk. Delve into the hardware structure and testing methods of these controllers, covering common architectures like Mobileye, Nvidia, TI, and Horizon. Examine software and hardware security risks, learn how to extract AI recognition models, and conduct security tests. Witness a demonstration of gaining control over a vehicle's autonomous driving domain controller to manipulate the entire vehicle. Gain insights into the often-overlooked security vulnerabilities of these crucial components, which utilize AI chips, high-precision GPS, multiple cameras, LiDAR, and other sensors for decision-making and vehicle control. Understand the potential dangers if hackers exploit these systems, emphasizing the importance of addressing these critical security issues in autonomous driving technology.
Overview
Syllabus
The Key to Remote Vehicle Control: Autonomous Driving Domain Controller
Taught by
Black Hat