The underestimated threat of cybersecurity to automated cars

Sunilchandra Dal
Saturday, 26 May 2018

The researchers say the cybersecurity threat to automated cars is underestimated and understudied. In a white paper titled ‘Assessing Risk: Identifying and Analysing Cybersecurity Threats to Automated Vehicles’ the researchers have given the following three interesting hypothetical situations that may arise due to cyberattacks:

When automated or self-driven cars were first introduced, they were projected as the mode of transport of the future. However, there is still time for these cars to win over the confidence of the general public. 

Safety of the occupants of the automated cars and other vehicles as well as pedestrians is a major issue. But other factors may arise as technology advances. Researchers at the University of Michigan, US, working with Mcity, are studying cybersecurity attacks on automated cars. Mcity, led by University of Michigan, is a public-private partnership to accelerate development of automated vehicles.

The researchers say the cybersecurity threat to automated cars is underestimated and understudied. In a white paper titled ‘Assessing Risk: Identifying and Analysing Cybersecurity Threats to Automated Vehicles’ the researchers have given the following three interesting hypothetical situations that may arise due to cyberattacks: 

“Instead of taking you home from work, your self-driving car delivers you to a desolate road, where it pulls off on the shoulder and stops.

You call your vehicle to pick you up from a store and instead you get a text message: Send $100 worth of Bitcoin to this account and it’ll be right over.

You buckle your seatbelt and set your destination to a doctor’s appointment, but your car won’t leave your driveway. It senses it’s been hacked and your home is its pre-programmed safe destination.”

These examples show the wide range of cybersecurity challenges that have to be faced before autonomous vehicles are widely accepted. 

The researchers have created a tool called Mcity Threat Identification Model to analyse the potential threats. The model outlines a framework for considering the attacker’s skill level; the vulnerable vehicle system components; the ways in which an attack could be achieved; and the repercussions on privacy, safety and financial loss. 

The paper says the automated cars will be vulnerable to data theft of personal and financial information, spoofers who present incorrect information to a vehicle, and denial-of-service attacks that move from shutting down computers to shutting down cars.

There are also security threats to networks that connect with autonomous vehicles including the financial networks that process tolls and parking payments, the roadway sensors, cameras and traffic signals, the electricity grid, and personal home networks. 

The researchers used the new tool to examine automated parking. A thief could spoof the remote parking signal to steal your car. This type of attack received a 7 out of 10 on the researchers’ scale of impact. The researchers noted that without robust, fool-proof cybersecurity for autonomous vehicles, systems and infrastructure, a viable, mass market for these vehicles simply won’t come into being.

Will networking enhance safety?

Writing for the Research University of Michigan website, Alex Piazza says driverless vehicle research underway at the University of Michigan may be able to cut down fatal mishaps, reduce fuel consumption and cut emissions. 

The new approach is to ‘connect’ vehicles to each other and surrounding infrastructure through wireless communication so they can anonymously and securely exchange data including location, speed and direction.

Thus, drivers and ‘cars’ can be warned of emerging dangerous situations ranging like road conditions ahead to vehicles approaching each other at a blind corner. 

Thus, networks of driverless vehicles connected with each other and the infrastructure can be managed, improving safety and traffic flow.  

Instead of using only current ‘smart’ cruise-control features, which respond to the car immediately in front, by connecting to a broader network of cars, the autonomous vehicles are able to slow down using less braking force than a human driver. 

The networked cars know what traffic is coming several cars away and so they react more efficiently. 

Related News