AI-powered Radar Advances Object Classification In Self-driving Cars


In order to allow autonomous vehicles make better sense of their surroundings, wireless technology startup Metawave has upgraded its automotive 77 GHz radar platform with a Micro Doppler Signature Extraction AI algorithm. The company’s WARLORD smart radar platform is now capable of distinguishing non-metallic objects like pedestrians, cyclists, and road animals like deer, squirrels, or coyotes, by analyzing their micro-motions.

The company’s radar platform uses Micro-Doppler Radar Signature along with various other AI features to recognize non-metallic objects, which the company says are difficult to spot through radar. In addition, Micro-Doppler Radar can also be used to provide additional information about metallic objects such as cars, motorcycles or trucks, as well as identify different classes of vehicles utilizing the Doppler signatures associated with the rotations of the wheels.

The company said that its Micro-Doppler Radar has been used in the military for target recognition, and there have been extensive academic studies from through-the-wall radar imaging, to classifying small UAVs to identifying human breathing patterns.


“When thinking about autonomous driving, we want to mimic two human fundamentals: perception and decision-making,” said Metawave CTO and co-founder Dr. Bernard Casse. “Metawave is building a superior radar ‘digital eye’ – possessing high-resolution 3D vision – and we’re also embedding decentralized intelligence, equipping the radar sensor with human-like interpretation of the world in order for the car to make good driving decisions.”

More information:

Renesas Develops Rapid and Precise Automotive AD Converter


Please enter your comment!
Please enter your name here

Are you human? *