Insurance Institute for Highway Safety, Ruckersville, Virginia.
Traffic Inj Prev. 2024;25(sup1):S78-S85. doi: 10.1080/15389588.2024.2374955. Epub 2024 Nov 1.
In 2021; half of crash fatalities occurred at night when some road users, like pedestrians, are particularly vulnerable. Automatic emergency braking (AEB) systems can avoid or mitigate collisions by automatically applying the brakes, but their performance may be hindered in low lighting. The purpose of this study was to estimate the proportion of real-world crashes where headlights could provide enough visibility for the driver or AEB system to detect and avoid the collision.
This study used IIHS headlight testing data and Crash Report Sampling System crash data from 2016 to 2021. The median 5-lux visibility distance was calculated for each IIHS headlight rating: good, acceptable, marginal, and poor. For three response profiles, a typical human driver, a fast human driver, and a camera-based AEB system, the maximum travel speed in which a vehicle could come to a complete stop within the 5-lux distance was computed. The real-world crash coverage of each headlight system and response profile was defined as the proportion of crashes that occurred on dark, unlit roads with a speed limit below the estimated maximum travel speed, where the vehicle could stop within the visibility distance.
The median low-beam visibility distance for poor-rated headlights was 68.2 m and for good-rated headlights was 103.1 m on the right side of straight roads. The typical human driver could have enough time to detect and avoid 46.2% of nighttime pedestrian crashes with poor-rated headlights and 70.1% with good-rated headlights. The visibility distance greatly increases with high beams and would allow the typical human driver to detect and avoid over 90% of nighttime crashes. Good-rated and acceptable-rated headlights allow AEB systems to theoretically have enough visibility distance to avoid nearly every nighttime crash.
Vehicles with good- or acceptable-rated headlights and active high beams are theoretically capable of providing sufficient light for speeds up to 165.8 km/h (103 mph), even covering most speeding vehicles. Future simulations of AEB systems on vehicles equipped with good- or acceptable-rated headlights and automatic high beams may not need to consider headlight visibility.
2021 年,有一半的撞车死亡事故发生在夜间,此时一些道路使用者(如行人)特别脆弱。自动紧急制动(AEB)系统可以通过自动刹车来避免或减轻碰撞,但在低光照条件下,它们的性能可能会受到影响。本研究的目的是估计现实世界中因车灯提供足够的能见度而使驾驶员或 AEB 系统能够检测到并避免碰撞的撞车事故比例。
本研究使用了 IIHS 车灯测试数据和 2016 年至 2021 年的 Crash Report Sampling System 碰撞数据。为每个 IIHS 车灯评级(良好、可接受、边缘和差)计算了 5 勒克斯能见度距离的中位数。对于三种响应配置文件(典型人类驾驶员、快速人类驾驶员和基于摄像头的 AEB 系统),计算了车辆在 5 勒克斯距离内完全停止的最大行驶速度。每个车灯系统和响应配置文件的现实世界碰撞覆盖范围定义为在黑暗、无照明且限速低于估计的最大行驶速度的道路上发生的、车辆可以在可见距离内停止的碰撞比例。
差评级车灯的右车道低光束能见度距离中位数为 68.2 米,良好评级车灯的能见度距离中位数为 103.1 米。典型的人类驾驶员可能有足够的时间来检测和避免 46.2%的夜间行人碰撞事故(差评级车灯)和 70.1%的夜间行人碰撞事故(良好评级车灯)。高光束大大增加了可见距离,这将使典型的人类驾驶员能够检测和避免超过 90%的夜间碰撞事故。良好评级和可接受评级的车灯使 AEB 系统在理论上有足够的可见距离来避免几乎所有夜间碰撞事故。
理论上,配备良好或可接受评级的车灯和自动远光灯的车辆能够以高达 165.8 公里/小时(103 英里/小时)的速度提供足够的光照,即使是覆盖大多数超速车辆。未来对配备良好或可接受评级的车灯和自动远光灯的车辆上的 AEB 系统的模拟可能不需要考虑车灯的可见度。