A security researcher has found an easy way to hack self-driving cars using a laser and a Raspberry Pi. By fooling the laser ranging (lidar) systems, hackers could make echoes of the fake car and put them at any location they want to fool the vehicle.
The technology companies are moving fast in the direction of bringing more and more intelligence to their products. Along similar lines, each week we hear new developments being made in the fields of automobiles and self-driving vehicles. Google has been testing its self-driving car with great excitement and sharing the pictures of the interiors of its cute-little self-driving car. However, the laser ranging (lidar) systems costing thousands of dollars could be hacked using a simple setup costing about $60.
Jonathan Petit, Principal Scientist at Security Innovation, says: “I can take echoes of a fake car and put them at any location I want. And I can do the same with a pedestrian or a wall.” In a research paper that will be presented at the Black Hat Europe in November, he tells how a pulse generator and a low-power laser can fool the car and play with its navigation.
Petit says: “It’s kind of a laser pointer, really. And you don’t need the pulse generator when you do the attack,” he says. “You can easily do it with a Raspberry Pi or an Arduino. It’s really off the shelf.”
Using this kind of system, a hacker can easily trick the self-driving cars into thinking that something is in front of it and thus slowing it down. This could also bombard tons of fake signals and car won’t move fearing of hitting the pedestrians and other fake cars. It should be noted that recently hackers hacked a jeep running at 70mph on highway
After studying the self-driving cars, Petit came to the conclusion that sensors used in them are the most vulnerable parts. He calls them the key point where input starts and if the car will have poor inputs, it’ll make dangerous decisions. The short-range radar systems for navigation deployed in these autonomous cars operate in a frequency band that needs licensing, whereas the lidar systems use hackable laser light to make a 3D picture of a car and surroundings.
IEEE writes: “Petit’s attack worked at distances up to 100 meters, in front, to the side or even behind the lidar being attacked and did not require him to target the lidar precisely with a narrow beam.”
It should be noted that Petit’s system didn’t utilize the vulnerabilities of the, instead it fooled them by mimicking the objects.
This latest car hacking tells us there are problems that need to be solved before allowing the self-driving cars to run in the wild. Security is always a big issue and the loopholes like this could be dangerous.
Having to something to add to this story? Tell us in the comments below.
Get your own raspberry pi hacker bundle here