[NetEase Smart News, August 28] As autonomous vehicles gradually become a reality on our roads, ensuring the safety of pedestrians is a growing concern. While there are numerous potential solutions, no definitive standard has been established yet.
At present, pedestrians have two primary ways to cross the road safely: either wait until there are no approaching vehicles or attempt to communicate directly with the driver. However, with the rising number of pedestrian deaths in the U.S., which increased by 25% since 2010 and reached over 5,000 in 2015, the advent of self-driving cars offers hope for improving this alarming safety record.
From a young age, we've been taught to make eye contact with drivers at crosswalks or intersections to ensure they see us. Yet, this is impossible with autonomous vehicles. Even if someone is seated in the driver's position, they are essentially passengers with little to no control over the vehicle’s actions. Often, these passengers could be distracted, working, watching movies, or even sleeping, completely oblivious to the situation ahead.
Alternative forms of communication between pedestrians and vehicles are now being explored. Various entities, including my own research group, tech companies, automakers, and startups, are investigating the use of visual signals to address this challenge—similar to a driver gesturing to a pedestrian or using headlights to indicate yielding. The complexity of achieving this is evident.
One issue is how people react upon recognizing an autonomous vehicle. Studies from institutions like the Virginia Tech Transportation Institute and the University of California, San Diego, show that disguising the actual driver helps simulate real-world scenarios where pedestrians might assume a car is unmanned. Similarly, Duke University's Human and Autonomous Systems Lab is focusing on how autonomous vehicles can effectively communicate with pedestrians.
[Image: A U.S. Standard "Stop Waiting" Signal]
Although there are federal guidelines for pedestrian traffic lights in transport systems today, there is currently no standard for vehicle-to-pedestrian communication. Establishing such standards is crucial. Implementing a universal crosswalk signal nationwide would enhance safety for both pedestrians and drivers alike. However, determining the most effective solution remains elusive.
Perhaps the indicator should be placed on the roof or front bumper—or somewhere in between—with a lit sign. But what message should it convey? Should it use multiple languages, or rely on universally recognized non-verbal symbols, akin to those at airports, making it comprehensible to anyone? Alternatively, should it emit sounds or music? Regardless of its form, the signal must be easily understood by individuals of all ages, educational backgrounds, and cultures.
On a two-lane road, displaying a simple text phrase like "Safe Passage" on a vehicle might suffice for pedestrians. However, scenarios become far more complex on multi-lane roads. For instance, can "Safe Passage" truly guarantee the absence of vehicles in adjacent lanes when crossing a four-lane road? What happens when multiple pedestrians approach from both sides of the street—how do they know which vehicle is safe to cross?
The design of these indicators is equally significant. To ensure readability from a distance of 100 feet—a current standard for fixed crosswalk signals—letters must be at least six inches tall. This requires a display nearly four feet wide. If the indicator is mounted on a moving vehicle, the size might need to be even larger.
Of course, the effectiveness of such indicators depends on visibility. In a study we conducted last year, we examined various display types mounted on vans disguised as unmanned vehicles. We found that only 12% of pedestrians used the displays to determine when it was safe to cross. Most still relied on traditional methods, such as gauging vehicle speed and distance to ensure adequate crossing time.
As the prevalence of human-driven cars diminishes, research labs like ours—alongside those in Virginia and California—will continue to investigate pedestrian responses to ensure everyone’s safety. These efforts mean more encounters with scientists disguised as drivers in test vehicles. (Translation by NetEase Intelligent Compiler Robot: Wu Man)
This article is brought to you by NetEase Smart Studio (WeChat ID: smartman163). Follow us for insights into AI and the next big technological era!
Pmic Led Drivers,Ic Pmic - Led Driver,Pmic Led Driver Ic Chip,Pmic Led Drivers Ic
Shenzhen Kaixuanye Technology Co., Ltd. , https://www.icoilne.com