The idea of unmanned vehicles filling the roads seems like a vision of the far future. However, with more and more automated features being part of our car, it may not be as far away as we think. We could expect fully automated cars on our roads as soon as 2020. While we welcome and look forward to technology that serves for making our lives easier, just how safe is this concept?

Principal scientist Jonathan Petit has taken it upon himself to carry out an experiment to show that automated vehicles may not be as safe as we thought. Security Innovation employee Petit works in the field of security for automated vehicles. He decided to test some of the automated features of cars that may one day replace drivers completely.

The Experiment

The Petit’s experiment took place in 2014. He focused on two automated features of vehicles: the cameras and the LIDAR system. LIDAR contains radar-like technology, using lasers to determine the distance between the vehicle and other bodies.

Currently, both these devices help the system make safe decisions as they drive. It is easy to imagine the damage that could happen if these elements could be externally tampered with.

The Results

Alarmingly, the camera proved incredibly easy to be intercepted. Petit found that simply shining a bright light into the lens would prevent its sensors from working effectively.

The LIDAR system did not prove much more difficult to hack either. By interrupting the laser before it received a response from the object it was trying to avoid, it was possible to trick the vehicle into thinking close objects were not actually there. The car would therefore lose its ability to identify its surroundings accurately.

Useful for Future Developments

The results of Petit’s experiment are valuable to developers of sensory technology for automated vehicles. While there is still a driver in control, these systems are useful in aiding their decision making and bringing to light things they may have been unaware of.

However, once the driver is removed and the decision making is left completely in the vehicle, safety becomes a lot more of a concern as there is no longer a human to override the decision if something goes wrong.

The fact that many of these systems can be so easily exploited must be addressed during the tests of prototypes of fully automated vehicles, so that they are much safer for common use.