基于人因理论的高风险交通行为多模态、差异化矫正技术数据集

数据集面向道路交通安全研究、我国智慧公路需求建设。基于人因理论设计了一种多模态的风险交通行为告警策略。基于长安大学汽车学院交通行业重点实验室的驾驶模拟器平台，开展了2组驾驶实验。在隧道典型风险场景中，开展了驾驶实验，共招募了48名被试参与者，获得了车辆运动数据、驾驶员操作数据、驾驶员眼动数据，数据量大小为4GB。在逆行车典型风险行模拟实验中，共招募了28名被试参与者，分别开展了逆行车与行人风险场景实验，实验获得了车辆运动数据、驾驶员操作数据、驾驶员眼动数据，数据量大小为7.23GB。

This dataset is developed for road traffic safety research and the construction demands of smart highways in China. A multimodal warning strategy for risky traffic behaviors was designed based on human factors theory. Two sets of driving experiments were conducted using the driving simulator platform at the Key Laboratory of Transportation Industry, School of Automobile, Chang'an University. For the driving experiment in the typical tunnel risk scenario, 48 participants were recruited, and vehicle motion data, driver operation data, and driver eye movement data were acquired, with a total data size of 4 GB. For the simulated experiment focusing on typical risk scenarios involving wrong-way vehicles, 28 participants were recruited, and experiments for two risk scenarios (wrong-way vehicles and pedestrians) were conducted respectively. The collected data include vehicle motion data, driver operation data, and driver eye movement data, with a total data size of 7.23 GB.

搜集汇总

背景与挑战

背景概述

该数据集基于人因理论，针对高风险交通行为设计了多模态矫正技术，包含隧道和逆行车典型风险场景下的驾驶实验数据，涵盖车辆运动、驾驶员操作和眼动数据，总数据量11.74GB，适用于道路交通安全研究和智慧公路建设。