Your vehicles are at risk of cybercrime. What can prevent it? (video)

Cybercrime has become a real threat to vehicles because of the increasing amount of software used in the latest vehicles. 

There are multiple points of attack in a car, including the OBD port, USB sockets, and that a connected vehicle’s access to the internet can also put it at risk. 

There has also been an incident where criminals accessed a vehicle’s software through its adaptive headlights. 

Outlining the risk at the September Fleet200 Strategy Network meeting was David Brown (pictured above), head of connected mobility product portfolio at Vodafone Automotive. 

Looking at fleet management specifically he highlighted three risk areas: 

1. Operational and personal data – an incident in the US saw data taken and held for ransom 
2. Vehicle takeover – a vehicle can be accessed remotely and “driven into a ditch” 
3. Sensor spoofing. Vehicle-generated data is intercepted and used for example in the making of fraudulent fuel claims. 

In response, Brown said, Vodafone takes a three-step approach – securing the device, the connection and the platform. 

Device: if the device is disassembled, the chipset is automatically destroyed. Plus rather than using a slide-in SIM card, like a mobile phone, its placed on the silicone and can’t be removed or replaced. Sensitive devices are also installed in mutiple difference places in a vehicle. 

Connection: Ensure the data and data flow is encrypted (AES 256), and “every device made in our factory it can only talk to our platform”, Brown said. 

A SIM card can only talk to one database endpoint, not all of the internet, which Brown described as the “Wild West”. This prevents a device communicating with non-legitimate endpoints. 

Platform: All Vodafone’s telematics’ data is stored on the same servers as its consumer data, with a “super high level” of protection, that includes 24/7 monitoring of global cyber threats. 

AI and the driver 

Vodafone uses machine learning in outward and driver facing cameras and can use AI to interpret what the driver is doing and determine their best course of action or highlight negative behaviour. Outward facing cameras can also read speed limit signs for example. 

Combined, this insight can be used for fuel optimisation and driver improvement, targeting those that need it the most. 

In the event of a crash, GPS data is collated pre and during the collision, together with accelerometer, g-forces information and what road signs have been passed to “potentially determine who was at fault”. 

Additional vehicle-related data can be used to offer predictive maintenance, for example on the condition of the catalytic converter, the diesel particulate filter, how often an EV is being charged to 100% versus 80%, Brown suggested. 

The information is combined using AI algorithms to help the fleet manager determine when a vehicle should be taken off the road for maintenance or actions taken to take care of the vehicle during its lifetime. 

“So, whilst connecting the car to the web can have its security risks, working with a provider that can secure its data, the outcomes are hugely beneficial,” Brown said. 

AI and natural language processing 

Vodafone has partnered with telematics and tracking company Geotab. It has introduced natural language processing enabling users to quiz the data using speech.  

“You can ask to see who your most dangerous drivers, or show you which vehicles need maintenance in the next month,” said Brown. “You don’t need a degree in reading through a huge amount of potentially mystifying data. The AI engine will quickly give you the answers.” 

Safer Transport for Europe 

The Safer Transport for Europe (STEP) programme is a European Vodafone investment and innovation programme using 5G technology. STEP sends information from vehicles and road furniture, such as signs and traffic lights, to a data collection point.  

Vodafone’s platform will then be used to inform road users, including cyclists and pedestrians, of traffic and safety issues.  

5G’s speed is crucial because it reduces the amount of time data can be sent, or pinged. 

“The best way to bring STEP to life is to give some use cases,” Brown said. In the event of heavy rain, STEP would warn you of road flooding before you reached it, the depth of the water based on the car that went through it in front of you, and therefore would give you guidance on whether you could go through it.” 

Another example is called time to green.  

“The typical behaviour driving through lights in a city involves lots of acceleration and braking as you go through green and stop at red lights. Time to green sees traffic lights providing feedback on changes, so you can amend your speed on approach to improve having to a complete halt or overly accelerating. 

“It improves driving efficiency and reduces vehicle emissions.” 

Another potential use case is displaying oncoming vehicle information in realtime on the back of a lorry trailer to a driver behind it, using a camera mounted on the front of a lorry to reduce the risk of overtaking accidents. 

STEP explained - launch video