论文部分内容阅读
为研究高速公路特长隧道下驾驶人视觉特征变化规律及其安全状态判别模型,要求被试驾驶人进行特长隧道环境的实车实验,同时利用眼动追踪装置采集驾驶人视觉特征参数。在统计分析数据的基础上,建立基于C4.5决策树的单一指标和综合指标判别模型。研究结果表明:相对于非隧道路段,隧道路段驾驶人瞳孔面积更大,注视时间更长;相对于入口段和出口段,行车段驾驶人瞳孔面积更大,而出口段则以短时注视为主;相对于出口段,入口段驾驶人瞳孔面积变化更大,注视持续时间更长;使用单一视觉参数指标可明显区分行车段和非隧道路段驾驶人安全状态,而基于双指标的识别模型可明显区分各路段驾驶人安全状态。研究进一步深化了高速公路特长隧道下驾驶人信息感知模型研究,为建立实时视觉特征判别安全状态提供理论基础。
In order to study the changing rule of the visual characteristics of drivers under the extra-long tunnel of expressway and the discriminant model of the safety state, the test driver is required to carry out the actual vehicle experiment of the tunnel with extra-long tunnel and the driver’s visual characteristic parameters are collected by eye tracking device. Based on the statistical analysis data, a single index and a comprehensive index discriminant model based on C4.5 decision tree are established. The results show that, compared with non-tunnel sections, driver’s pupils in tunnel section have larger area and longer fixation time; driver’s pupil area is larger than that in entrance section and exit section, while exit section is short-term Main; relative to the exit section, the entrance section of the driver’s pupil area changes more, watching longer duration; using a single visual parameter indicator can clearly distinguish between driving section and non-tunnel section driver safety status, and based on two-index identification model Clearly distinguish the safety of drivers in all sections. The research further deepens the study of driver information perception model under the expressway super long tunnel, and provides a theoretical basis for establishing the real-time visual characteristics to identify the security status.