The proliferation of autonomous driving in recent years has led to valid concerns about safety on the road for pedestrians as well as other drivers. With brands like Nissan and Volkswagen incorporating self-driving technology into their vehicles and ride share apps like Uber and Waymo transporting passengers in driverless vehicles, it seems like the advancement is here to stay.
Now, with several years of data at our disposal, a group of researchers sought to determine if self-driving cars were actually more dangerous than those driven by humans. They just published their findings in the Nature journal.
The scientists examined crash data from 2,100 autonomous vehicles and 35,113 human-driven vehicles in California from 2016 to 2022, as the Golden State has been a popular testing ground for autonomous vehicles in recent years. They found that self-driving vehicles had lower accident rates than humans in construction zones, traffic events, and certain "pre-accident" movements like slowing down and moving forward. They even deduced that autonomous vehicles drove safer in rainy conditions than humans did.
There are a few catches, though. The self-driving cars underperformed greatly depending on the time of day and what routes they were driving. Autonomous driving at sunrise and sunset was particularly hampered.
"This could be attributed to the sensors and cameras used by [autonomous vehicles] may not be able to quickly adapt to changes in lighting conditions, which could affect their ability to detect obstacles, pedestrians, and other vehicles," the team wrote in the paper. "At dawn and dusk, for instance, the sun’s shadows and reflections may confuse sensors, making it hard for them to distinguish between objects and identify potential hazards."
"Furthermore, the fluctuating light conditions can impact the accuracy of object detection and recognition algorithms used by [autonomous vehicles]," they added.
Related: Watch Angry Resident Smash Up Self-Driving Taxi With Hammer
Researchers also said that not every driving situation is simple enough for a computer to figure out, as many times the quick thinking of a human can be what prevents an accident. The team noted the danger that "multiple oncoming [human-driven vehicles]" or "unprotected left turns at intersections," for example, could present.
Plus, pedestrians and bikers have already been mowed down by self-driving cars before—an advantage that humans can still easily claim.
"[Human-driven vehicles] can predict pedestrian movements and exercise caution based on their driving experience, whereas [autonomous vehicles] may struggle with recognizing pedestrians’ intentions, potentially leading to emergency braking or accidents due to a lack of understanding of social cues and psychological reasoning," the researchers claimed.
This information could come in handy as car companies continue to develop and improve the autonomous driving technologies in their vehicles. Still, just because there might be less humans behind the wheel doesn't mean road rage will decrease.
from Men's Journal https://ift.tt/5zGje0Q
via IFTTT