为进一步揭示高速公路交通动态交通参数与交通冲突事件的关系.利用highD数据库以3 min为基本单位构建样本,共提取交通流和路段特征23个.基于交通冲突衡量指标后侵入时间(post encroachment time,PET)和不同阈值,统计车辆跟驰和变道过程中不同严重程度的冲突数量.基于随机森林回归(random forest regression,RFR)进行特征筛选,利用数值特征灰度图片转换技术(feature matrix to gray image,FM2GI)技术将各样本转化为灰度图片,结合2D-卷积提取图片特征,构建了卷积神经网络(convolutional neural net-works,CNN),C-RFR,C-SVR这3个编码-解码模型,并与基准模型BP神经网络(back propagation neuralnet-work,BPNN)、RFR、支持向量回归(support vector regression,SVR)对比分析.结果表明,基于路段驶入车辆数和驶出车辆数等2个重要特征,平均车头时距、时间占有率、客车平均行驶速度、换道率和驶出点速度方差等5个有效特征,编码-解码架构下的CNN、C-RFR和C-SVR均优于直接应用基准模型,均方根误差(root mean squared error,RMSE)分别降低了 12.6％,31.6％,18.5％,能够实现交通冲突的实时预测.其中CNN预测误差最小,且在应对不同严重程度的交通冲突预测时表现出良好的鲁棒性,及对2个关键参数表现出低敏感性.基于FM2GI技术和2D-卷积编码的CNN、C-RFR和C-SVR模型拓展了交通冲突预测建模的深度学习框架,可实现高速公路基本路段多严重程度交通冲突的可靠预测.
Deep Learning Prediction of Expressway Traffic Conflicts Based on The Encoder-Decoder Architecture
The approach aims to uncover the relationship between dynamic traffic parameters and traffic conflict in-cidents and it further supports proactive safety control.The highD database is utilized to create sample data in 3-minute intervals,extracting 23 features related to traffic flow and road section characteristics.Based on the post encroachment time(PET)indicator,different thresholds are set to classify the severity of conflicts during car-fol-lowing and lane-changing scenarios.The random forest regression(RFR)method is used to select the most criti-cal features,while feature matrix to gray image(FM2GI)technology converts the sample data into grayscale im-ages to enable 2D convolution to extract image features.Three encoder-decoder models,convolutional neural net-works(CNN),C-RFR,and C-SVR are compared with baseline models(back propagation neural network(BPNN),RFR,and support vector regression(SVR).The results indicated that:based on two key features(the number of vehicles entering and exiting the road section)and five effective features(average headway,time oc-cupancy,average driving speed of passenger cars,lane change rate,and variance of exit speeds),the CNN,C-RFR,and C-SVR models within the encoder-decoder framework outperformed the baseline models.Spe-cifically,root mean squared error(RMSE)reduced by 12.6％,31.6％,and 18.5％,respectively,enabling re-al-time prediction of traffic conflicts.Among them,CNN exhibited the lowest prediction error and demonstrated strong robustness in predicting traffic conflicts of varying severities,along with low sensitivity to two key param-eters.The CNN,C-RFR,and C-SVR models,utilizing FM2GI technology and 2D convolution encoding,expand the deep learning framework for traffic conflict prediction modeling,and achieve reliable predictions for multiple severities of highway traffic conflicts in basic road segments.
万方数据
维普
