Speed cameras are already widely used in cars to help with emergency braking, blind spot detection and parking assistance systems, but low image resolution limits its use, said Steven Hong, CEO from the Dayton, Ohio startup.
Artificial intelligence software developed by Oculii can increase radar resolution by 100 times, Hong said. The image created by the software and presented to Reuters resembled images and maps created using lidars, a laser sensor used in self-driving cars.
“Our particular approach leverages existing, market-proven hardware that has been deployed at very low cost,” Hong said. “With software, it is able to achieve lidar-like resolutions, improving and evolving rapidly to one day exceed the performance of these lidar sensors.”
Oculii licenses the software to automotive suppliers to integrate speed cameras into systems such as Advanced Driver Assistance Systems (ADAS), although Hong said he could not divulge the names of customers or partners. .
While lidar prices have fallen rapidly, they are too expensive for use in mass-produced products, including cars. Still, several self-driving car tech company executives told Reuters they believe lidars will be necessary for the future of the robotaxi, even with improved radar.
Discover the stories that interest you
Hong said Oculii is testing with some self-driving car companies in China and North America using only radar and cameras. He said Oculii also sells kits that can be integrated into autonomous robots and drones.
Oculii’s latest funding round was co-led by Catapult Ventures and Conductive Ventures, but Oculii declined to disclose its latest valuation.
Hong co-founded the company with his father Lang Hong, who has worked on radar for three decades as a professor and dean of the school of engineering at Wright State University.