In recent years, technological advances have made it possible to envisage the best for autonomous cars. Since June 1, 2018, our customer Transports Publics Genevois (TPG) has obtained official authorization to launch an automated line of automated vehicles for Geneva commuters. These advances highlight cybersecurity for autonomous cars.
Why has TPG launched an autonomous car project?
To improve traffic flow in the Geneva area and reduce greenhouse gas emissions, TPG is introducing autonomous vehicles to enable users to travel at lower cost while emitting zero carbon emissions. To bring this major project to fruition, Transports Publics Genevois had to invest 20 million euros to make it viable and safe in every respect.
Cybersecurity, a little-known issue for autonomous car users
The safety of the user and the surrounding environment, especially other motorists, is the first question on everyone's mind, and rightly so. What is less obvious to everyone, however, are the cybersecurity issues associated with these autonomous vehicles. Autonomous vehicles using non-reserved lanes, and therefore usable by other motorists, need to be managed from both a "human safety" and a "cybersecurity" point of view.
Thanks to the NioVe project, a pilot project prior to the release of the Navia project, our teams of experts Team Partners were able to improve the cybersecurity of autonomous vehicles by conducting a battery of tests.
How can we ensure the cybersecurity of autonomous vehicles?
Our experts reproduced the real-life environment in which the vehicle travels. During our tests, we installed cameras in a streetcar to be able to carry out attacks on the vehicle's positioning. We had to list and understand the various cyberattacks faced by the vehicles.
In particular, attacks on the Global Navigation Satellite Systems, i.e. attacks on the vehicle's positioning. So we tricked the vehicle, gave it false locations and tried to detect its faults.
We also tested attacks on the entire video section to ensure the cybersecurity of autonomous vehicles. We installed cameras on a streetcar and carried out cyberattacks on these videos. On the video input side, we carried out a number of different attacks. We changed the colors and added artifacts to see if we could detect any manipulation of the image.
Test results
Thanks to our research and testing, we've been able to try and get vehicles into trouble and see how it detects these attacks. To prevent a vehicle from changing its position and no longer understanding where it is, the tests we carried out to try and lose it were not successful. The vehicle was able to counter our attacks and stay on course despite our false instructions.
Thanks to these tests, vehicles have been taught to detect, understand and anticipate risks. Prevention has therefore proved to be the key to increasing the analysis capacity of autonomous vehicles and ensuring their cybersecurity.
Support Team Partners
Our teams at Team Partners teams are helping Transports Publics Genevois (TPG) to meet the cybersecurity challenges of their autonomous vehicles. In this new era of autonomous vehicles, the safety of local residents, other vehicles and users is a key issue.
