基于近红外技术的道路线形认知实验研究  

Natural Driving Experiment on Road Alignment Cognition Based on Near Infrared Technology

作　　者：赵微[1] 陈雨人[1] ZHAO Wei;CHEN Yuren(The Key Laboratory of Road and Traffic Engineering,Ministry of Education,Shanghai 201800,China)

机构地区：[1]同济大学道路与交通工程教育部重点实验室,上海201800

出　　处：《交通与运输》2025年第1期72-75,共4页Traffic & Transportation

摘　　要：为深入了解驾驶人在驾驶过程中对于道路线形的认知活动,通过功能性近红外光谱技术观测驾驶人前额叶皮层中氧合血红蛋白浓度变化开展自然驾驶实验研究:通过移动平均滤波器对前额叶皮层区域的4个通道信号进行滤波平滑处理;选择CatMull-Rom三次样条曲线对视觉车道形态进行拟合并划分视觉车道区域,选取不同视野景深范围的曲线长度和视觉速度作为视觉车道线形特征,并进一步通过LightGBM算法进行数据建模。模型预测结果表明,左右两侧的道路近景曲线长度和左侧道路中景曲线长度对通道区域附近的驾驶人前额叶皮层中氧合血红蛋白浓度的影响较大,是影响驾驶人认知活动与执行功能最重要的道路线形特征。In order to gain a deeper understanding of the driver's cognitive activities on the road alignment during driving,a natural driving experiment was conducted and the changes in the concentration of oxygenated hemoglobin in the driver's prefrontal cortex were observed using functional near-infrared spectroscopy.The four channel signals in the prefrontal cortex area were filtered and smoothed using a moving average filter;the CatMull-Rom cubic spline curve was selected to fit the visual lane morphology and divide the visual lane area.The curve length and visual speed in different depths of field were selected as the visual lane alignment features,and the data was further modeled using the Light GBM algorithm.The model prediction results show that the length of the near-view curve on the left and right sides of the road and the length of the mid-view curve on the left side have a greater impact on the concentration of oxygenated hemoglobin in the prefrontal cortex near the channel area,and are the most important road alignment features that affect the driver's cognitive activities and executive functions.

关 键 词：驾驶人认知 近红外光谱技术 样条曲线拟合 LightGBM算法 道路安全 

分 类 号：U491[交通运输工程—交通运输规划与管理]