100-Car Naturalistic Driving Study (NDS)

该数据集名为100-Car NDS，是首个在北弗吉尼亚/华盛顿特区地区开展的为期两年的大型车载仪器研究，旨在收集大规模的自然驾驶数据。该数据集包含了关于碰撞和险些碰撞的详细信息，视频资料能够用于分析事件前和事件中的参数。此外，它还提供了总共57个变量来描述驾驶事件。数据规模约为200万车辆里程、43,000小时的数据记录以及102名主要驾驶员。该数据集的任务是针对驾驶场景中的碰撞缓解和避免进行研究。

This dataset, named 100-Car NDS, is the first large-scale instrumented in-vehicle study conducted over a two-year period in the Northern Virginia/Washington, D.C. area, with the goal of collecting large-scale naturalistic driving data. It contains detailed information on crashes and near-crashes, and video footage is available for analyzing pre-event and in-event parameters. Additionally, it provides a total of 57 variables for describing driving events. The dataset encompasses approximately 2 million vehicle miles, 43,000 hours of recorded data, and involves 102 primary drivers. This dataset is intended for research on crash mitigation and avoidance in driving scenarios.

搜集汇总

背景与挑战

背景概述

该数据集是一个大规模自然驾驶行为研究，包含约100辆装备多种传感器的车辆在两年内收集的约200万英里和43,000小时的驾驶数据，特别包括68次碰撞和760次接近碰撞的详细视频分析。数据集旨在通过非侵入式仪器和无特殊指令的参与者，真实记录驾驶行为和环境因素。