Autonomous and unmanned systems are already patrolling our skies and oceans and being tested on our streets and highways. Pioneering tests of autonomous vehicles were performed in Europe and all trends indicate these systems are at an inflection point that will show them rapidly becoming commonplace. This presentation will discuss the capabilities and potential vulnerabilities of these systems, and summarize recent research into successfully subverting various components.
Academic teams, tech startups and automotive companies are putting high-profile efforts into driverless cars, and gaining plenty of headlines. But autonomous and unmanned systems are already patrolling the skies and oceans in addition to being tested on our streets and highways. It’s already past time to be discussing the security implications and vulnerabilities of these machines which operate with delayed, remote supervisory control or without a human in the loop at all. The USA gets most of the driverless vehicle attention, but pioneering testing of autonomous ground vehicles is currently being performed in Europe, especially in Germany. All trends indicate that these systems collectively (including supervised driver-assist modes) are at an inflection point that will bring them rapidly into common usage. Security researchers and professionals therefore need to be aware of the capabilities and vulnerabilities of these systems and their components.
In this session, a researcher that’s been intimately involved with the autonomous and unmanned systems community for many years will walk you through an informative and at times amusing look at the current state of civil driverless vehicles and what researchers, hackers and criminals can and have done to subvert them. You’ll be brought up to speed with common sensors, decision systems, failure modes and current adversarial research. The talk will also cover proposed protocols for driver assistance, such as Tesla AutoPilot, and automated vehicle safety, through the US Department of Transportation’s new Connected Vehicle and Vehicle To Vehicle Communications programs, that are likely to be adopted earlier than fully autonomous cars but bring their own unique vulnerabilities onto the road.