交通环境及驾驶人属性对车辆横向控制水平的影响分析

Influence of the environment and driver attributes on vehicle lateral control level

为探究道路环境及驾驶人属性对车辆直道横向控制水平的影响,招募31名试验对象,设计白天-城市道路、夜晚-城市道路、白天-高速公路、夜晚-高速公路等4种情境下的实际道路驾驶试验,选取横向加速度均值和横向加速度标准差为二维向量,基于二分K-means算法对驾驶人在直道行驶的横向控制水平进行划分。结果表明:样本数据聚为3类时聚类效果最佳,驾驶人的直道横向控制水平被划分为良好、一般、较差;4种情境中,夜晚-城市道路环境下驾驶人的整体横向控制水平最差;所有驾驶人属性中,只有性别因素对驾驶人的横向控制水平存在显著影响,男性驾驶人的直道横向控制水平优于女性驾驶人,年龄、驾龄、婚姻状况、文化程度等5种因素不存在显著影响。

The lateral control ability of drivers is especially important for lane keeping and smooth transitions.To investigate the effect of coupled sunlight conditions and road types on the lateral control level of drivers'vehicles and the differences between drivers with different individual attributes,this paper designed a real road driving experiment under four scenarios,including daytime-city road,nighttime-city road,daytimehighway,and nighttime-highway.Thirty-one subjects were recruited,and the Volkswagen Golf new energy vehicle was used as the experimental vehicle to collect the vehicle driving status data during the driving process with the help of acceleration sensors,three-in-one multifunctional modules,and GPS Beidou external antennas.Based on the dichotomous K-means algorithm,the driver's vehicle lateral control level in different traffic environments is divided,and the number of clusters is determined by combining the contour coefficient method.At the same time,basic information such as driver's name,gender,age,education level,driving age,marital status,and sleep time were collected to analyze the influence of driver attributes on the lateral control level of the vehicle.The results show that with the help of the dichotomous K-means algorithm to classify the lateral control level of drivers,the samples are highly differentiated and have a good clustering effect.The contour coefficient value reaches the maximum when the sample data are clustered into 3 classes,and the clustering effect is the best.The lateral control level of drivers on straight roads is classified as good,fair,and poor;among the 4 contexts,the overall lateral control level of drivers in night-urban road environment is the worst;among all driver attributes,only the gender factor has a significant effect on the lateral control level of drivers driving on straight roads,and the male drivers'straight.The lateral control level of male drivers is better than that of female drivers,and there is no significant effect of five factors,such as age,driving age,marital status,and education level.