MIT develops ground-penetrating radar to help driverless cars navigate in bad weather

Self-driving technology has come a long way, but it can still be hampered by bad weather. A team at MIT's Computer Science and Artificial Intelligence Laboratory may have a solution. They've developed a method that helps vehicles "see" ground that's obscured by the environment by mapping the road below using ground-penetrating radar (GPR).

MIT test car (Photo source: Engadget)

Most self-driving vehicles use lidar sensors or cameras to see around them, but cameras can be blocked by lighting conditions or snow cover and cannot see traffic signs or road conditions on the ground. Lidar tends to be less accurate in bad weather. Ground-penetrating radar sends electromagnetic pulses to the ground to measure the specific composition of soil, rocks and other materials. These data are converted into driving maps for self-driving vehicles.

So far, the team has only conducted low-speed tests on closed country roads, but the researchers believe it could easily be expanded to highways and other high-speed areas. They admit that the system doesn't work when rainwater seeps into the soil beneath the road surface, and it's far from road-ready. It also has to be used in conjunction with other technologies, otherwise it won't sense danger while driving.