Scholarlink

<p>Intersection accidents primarily involve left-turning vehicles facing oncoming traffic and vehicles crossing paths. Many car manufacturers have implemented advanced safety features such as automated emergency braking (AEB) systems to reduce these incidents, although such technologies are mostly installed in high-end vehicles. This study evaluates the reaction time and sensor technology of AEB systems based on real accident data to analyze the likelihood of collisions at intersections. It examines how the field of view (FOV) of sensors and braking delay times impact collision frequency across different vehicle models. The research compares a path prediction model with a path intrusion model. The path prediction model effectively prevents collisions, especially in scenarios where vehicles continue to drive straight. Additionally, the wider the FOV of the sensors and the shorter the braking delay, the higher the collision prevention efficiency. Systems with less than 0.1 seconds of braking delay and a wide FOV are particularly effective in preventing collisions in complex environments like intersections. This study recommends that automotive manufacturers and technology developers enhance vehicle safety features by widening the sensor’s FOV and minimizing braking times. Such technical improvements can not only prevent collisions at intersections but also enhance overall vehicle safety.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Accident prediction considering sensor delay and FOV changes on intersections

  • 期刊论文

  • 发表日期：

    2025年5月6日

查看全文
  • Yunsik Shin,
  • Myungyeun Park,
  • Jayil Jeong

摘要

Intersection accidents primarily involve left-turning vehicles facing oncoming traffic and vehicles crossing paths. Many car manufacturers have implemented advanced safety features such as automated emergency braking (AEB) systems to reduce these incidents, although such technologies are mostly installed in high-end vehicles. This study evaluates the reaction time and sensor technology of AEB systems based on real accident data to analyze the likelihood of collisions at intersections. It examines how the field of view (FOV) of sensors and braking delay times impact collision frequency across different vehicle models. The research compares a path prediction model with a path intrusion model. The path prediction model effectively prevents collisions, especially in scenarios where vehicles continue to drive straight. Additionally, the wider the FOV of the sensors and the shorter the braking delay, the higher the collision prevention efficiency. Systems with less than 0.1 seconds of braking delay and a wide FOV are particularly effective in preventing collisions in complex environments like intersections. This study recommends that automotive manufacturers and technology developers enhance vehicle safety features by widening the sensor’s FOV and minimizing braking times. Such technical improvements can not only prevent collisions at intersections but also enhance overall vehicle safety.