HARMAN International, a demonstrated at the CES in Las Vegas new detection capabilities, part of the HARMAN SHIELD Solution, that protect autonomous and semi-autonomous vehicles against cyber-attacks aimed at the vehicle’s sensors.
All HARMAN SHIELD solutions comprises of leading building blocks to provide a modular scalable architecture. HARMAN SHIELD key components include:
In-vehicle agents – embedded on and protecting key entry-points to critical assets, including TCUs, Head units and central gateways. SHIELD agents all employ HARMAN award-wining IDS technology and are equipped with full backend reporting capabilities and over-the-air updatability.
SHIELD smart client – a centralized client-side orchestrator to provide smart aggregation of data and standard reporting to the backend.
Cybersecurity Analysis Center – a full dashboard and analytics solution providing 24/7 visibility of broad vehicular security related events from HARMAN SHIELD Agents, integrated on-top of HARMAN Ignite Platform.
API for 3rd party integrations – Import HARMAN SHIELD data into Security Operation Center (SOC), Security Incident and Event Management (SIEM) solutions – such as IBM QRadar and others.
The technology demonstration shows how an adversary image with a speed limit sign was able to fool the on-board traffic sign recognition system of a production vehicle, to present false information to the driver and impact other vehicle systems, such as adaptive cruise control. That spoofed traffic sign was successfully detected and reported to HARMAN Cybersecurity Analysis Center – a full dashboard and analytics solution which provides 24/7 visibility of broad vehicular security-related events from HARMAN SHIELD Agents.
Adversary images are images that have been intentionally manipulated in such a way that a human recognizes them correctly but a neural network or computer vision-based system will misclassify them. The use of such images opens a way to new attacks on autonomous cars and ADAS systems, requiring non-conventional cybersecurity techniques to detect and possibly mitigate. These attacks do not require any physical access to the car or tampering with the communication system and the consequences of such attacks can be grave, including major traffic disruptions.
“In 2017, dozens of cities around the world have deployed autonomous vehicle pilots, from San Francisco to Las Vegas and London, all while the attack surface of autonomous vehicles continues to grow and change,” said Yuval Weisglass, vice president, automotive cyber security at HARMAN Connected Services. “In order to protect autonomous vehicles against these types of cyber-attacks, now is the time to adopt a security-by-design approach, developing unconventional detection and protection capabilities. As part of our ongoing investment in R&D, we constantly revalidate our threat assessments in order to identify new fields that might impact the attack surface of connected and autonomous vehicles.”
More than 30 cities around the world are already piloting autonomous vehicles projects, with nearly 100,000 Fully Autonomous vehicles expected to enter the market in 2020 – vehicles which can be hacked by attacking their sensors or manipulating the environment these vehicles interact with. In addition, IHS Markit forecasts that 25% of vehicles sold globally 6 years from now will be equipped with cybersecurity cloud services, such as those offered by HARMAN. “Cybersecurity is becoming a key technology for the automotive industry as connected cars grow,” said Egil Juliussen, Ph.D. and research director for IHS Markit. “It is especially important for self-driving and driverless cars where it will be required.”