Princeton Researchers Use AI To Create Radar That Sees Around Corners
Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!
Tesla provided the first clues that radar could be trained to do more than just detect objects straight ahead. After the death of Joshua Brown on a Florida highway in 2016, Tesla tore up the Autopilot software created by MobilEye and pivoted from a camera-based to a radar-based system. In the process, it learned how to bounce radar signals under the car directly ahead to “see” what the next car in line was doing. That way, if a truck or SUV is blocking the view of the road ahead, a Tesla with the updated system could still detect if a car further up the road slowed or braked unexpectedly and take appropriate action.
One of the challenges of designing control systems for autonomous driving cars is anticipating if a pedestrian, bicyclist, or other motorist is about to cross the intended path of the vehicle. Various high tech (and high cost) systems have been tried with only limited success.
Now researchers at the Princeton School of Engineering say they have created an artificial intelligence algorithm that can tease that information out of the background noise associated with ordinary — an inexpensive — radar units. Their research was presented to the Conference on Computer Vision and Pattern Recognition on June 16.
“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” says Felix Heide, an assistant professor of computer science at Princeton and one of researchers. “The radar sensors are also relatively low-cost, especially compared to lidar sensors, and scale to mass production.”
According to an online post by Princeton,
“The system, easily integrated into today’s vehicles, uses Doppler radar to bounce radio waves off surfaces such as buildings and parked automobiles. The radar signal hits the surface at an angle, so its reflection rebounds off like a cue ball hitting the wall of a pool table. The signal goes on to strike objects hidden around the corner.
“Some of the radar signal bounces back to detectors mounted on the car, allowing the system to see objects around the corner and tell whether they are moving or stationary. 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 algorithms that we developed are highly efficient and fit on current generation automotive hardware systems,” Heide says. “So, you might see this technology already in the next generation of vehicles.”
To permit the system to distinguish objects not visible to optical sensors such as cameras, Heide’s team processed part of the radar signal that standard radars consider background noise rather than usable information. The team applied artificial intelligence techniques to refine the processing and read the images. Fangyin Wei, a graduate student in computer science and one of the paper’s lead authors, said the computer running the system had to learn to recognize cyclists and pedestrians from a very sparse amount of data.
“First we have to detect if something is there. If there is something there, is it important? Is it a cyclist or a pedestrian?” she said. “Then we have to locate it.” Wei said the system currently detects pedestrians and cyclists because the engineers felt those were the most challenging objects because of their small size and varied shape and motion. She said the system could be adjusted to detect vehicles as well.
Since the Princeton system relies on existing radar sensor technology, incorporating it into the next generation of automobiles should be a straightforward process, Heide says. “In terms of integration and bringing it to market, it requires a lot of engineering. But the technology is there, so there is the potential for seeing this very soon in vehicles.”
The research team included Jürgen Dickmann, Florian Krause, Werner Ritter, and Nicolas Schiener of Mercedes-Benz, Buu Phan and Fahim Mannan of Algolux, Klaus Dietmayer of Ulm University, and Bernard Sick of the University of Kassel. Partial funding was provided by the European Union’s H2020 ECSEL program.
Training radar to see around corners could be an important step forward in automobile safety and substantially reduce injuries to people walking or riding bikes in the vicinity of vehicles equipped with this new technology.
Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!
Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.
Sign up for our daily newsletter for 15 new cleantech stories a day. Or sign up for our weekly one if daily is too frequent.
CleanTechnica uses affiliate links. See our policy here.
CleanTechnica's Comment Policy
