“We face a crisis on America’s roadways that we must address together,” said U.S. Transportation Secretary Pete Buttigieg. “With our National Roadway Safety Strategy and the President’s Bipartisan Infrastructure Law, we are taking critical steps to help reverse this devastating trend and save lives on our roadways.” according to a study released last May.
NHTSA projects that an estimated 42,915 people died in motor vehicle traffic crashes last year, a 10.5% increase from the 38,824 fatalities in 2020.
Further, a recent report [1] from the Insurance Institute for Highway Safety (IIHS) reached an alarming conclusion. Not only have pedestrian injuries and deaths been rising for years but the systems now installed on automobiles intended to protect pedestrians, fail to work at night when more than 76% of the fatalities occur.
Read that again - the systems intended to protect pedestrians fail at night.
This realization has prompted changes in the automobile safety test protocols and intensified the search for reliable methods of classifying pedestrians at night.
Discussions on the need to improve pedestrian safety have been underway since before the first pedestrian crossing signal was installed in England in 1868 and continues to this day. But only recently, in 2010, national traffic safety regulators mandated the first active system directed at pedestrian safety.
By then, the computers in cars were powerful enough to manage signals from a variety of sensors and building vehicles that could operate without drivers became a desirable goal. Distance sensors using radar provided the basis for vehicle- to-vehicle anti-collision systems, raising the hope that if these sensors could accurately locate pedestrians and other VRUs, collisions with pedestrians and cyclists might be reduced.
FAILURES OF CURRENT SYSTEMS
Protecting pedestrians from collisions with cars in the maximum variety of situations requires the use of pedestrian automatic emergency braking (PAEB) systems that do not depend on the reaction of a driver to perceive a pedestrian in the car’s path. While the existing systems using cameras and radar operate well during the day when visibility is good, they fail when vision is obscured by atmospheric conditions, at night in the dark, in the presence of bright backlighting, or in chaotic urban scenes, as depicted in these examples.
This is where Owl AI comes in. Owl AI has recently announced a high-definition 3D Monocular Thermal Ranger(TM) system that leans in to the problems associated with protecting pedestrians and other vulnerable road users (VRUs) at night, and calls on thermal imaging and multimodal fusion to deploy improvements to current automotive safety protocols and reduce fatalities after dark (and, other comprised visibility environments.)
Download our latest white paper - New Regulations for Cars - to Protect Pedestrians at Night to read the full story.