New Doppler Radar System Can Detect Moving Vehicles Around Corners

Doppler radar systems, which are typically used to detect speeding drivers, are relatively low cost.
Chris Young

A new radar system has been created by scientists at Princeton University that allows automobiles to detect other moving vehicles and pedestrians around corners in order to avoid collisions.

The system, which is easily integrated into today's vehicles, uses Doppler radar to bounce radio waves off surfaces such as buildings and parked automobiles. It allows a level of collision detection that simply wouldn't be possible for humans.

RELATED: FUJITSU'S AI PORT IN JAPAN REDUCES SHIP COLLISIONS

Seeing around dangerous intersections

Doppler radar's signal hits the surface at an angle, meaning its reflection rebounds off a surface like a cue ball hitting the wall of a pool table, a Princeton press release explains.

The signal will then go on to strike objects hidden around the corner. Some of these signals bounce back and can be picked up by detectors mounted on a moving car, effectively allowing the system to see objects around the corner and tell whether they are moving or stationary.

New Doppler Radar System Can Detect Moving Vehicles Around Corners
Source: Princeton University

"This will enable cars to see occluded objects that today’s lidar and camera sensors cannot record, for example, allowing a self-driving vehicle to see around a dangerous intersection," Felix Heide, an assistant professor of computer science at Princeton University and one of the researchers, explained.

"The radar sensors are also relatively low-cost, especially compared to lidar sensors, and scale to mass production," he continued.

Making the roads of the future safe

In a paper presented on June 16 at the Conference on Computer Vision and Pattern Recognition (CVPR), the Princeton researchers detailed how their system is able to detect moving objects — including bicycles, pedestrians, and cars — and determine their direction and oncoming speed.

"The proposed approach allows for collision warning for pedestrians and cyclists in real-world autonomous driving scenarios — before seeing them with existing direct line-of-sight sensors,” the authors wrote in their paper.

The findings could prove to serve as another string in the bow for existing systems, such as Tesla's Autopilot. Ultimately, it is part of a concerted effort to make the roads of the future much safer than they are now.

Add Interesting Engineering to your Google News feed.
Add Interesting Engineering to your Google News feed.
message circleSHOW COMMENT (1)chevron
Job Board