基于有序Logistic回归的驾驶人危险辨识影响因素分析

Influencing Factors Analysis of Driver’s Hazard Perception Based on Ordered Logistic Regression

针对驾驶人危险辨识影响因素,开展危险辨识试验数据与自我报告数据的采集,采用k-means聚类分析方法将危险辨识时间划分为3类,进而基于有序Logistic回归分析方法检验不同影响因素的显著性。试验分析结果表明,危险情境特征对驾驶人危险辨识时间有影响,交通情境越混乱,所需时间越长,且交通情境越危险,所需时间相对越短;同时,驾驶人特征及其经历对驾驶人危险辨识时间有影响,男性的危险辨识时间较女性更短;摩托、电动车驾驶经历越丰富,危险辨识能力相对越好;驾龄、驾驶里程与自行车驾驶经历对危险辨识时间并无显著影响。环境因素和人口社会学因素对驾驶人危险辨识能力的影响是多方面的。

Aiming at the influencing factors of hazard perception of driver, a hazard perception experiment and self-reported data collection were carried out. The hazard perception time was divided into three categories using k-means clustering analysis, and the ordered Logistic regression analysis method was used to test the significance of the different factors. The results of experiments show that the characteristics of dangerous condition have an impact on hazard perception of driver. The more complicate the traffic condition is, the longer the time is required, and the more dangerous the traffic condition is, the shorter the time is required. Meanwhile, the characteristics and experience of driver affecte the hazard perception time. The time of male driver is shorter than that of female one. The richer the driving experience of motorcycles or electric bicycles is, the better the hazard perception. However, the number of years of driving, the mileage, and the bicycle driving experience show no significant effect on the hazard perception. Therefore, the influence of environmental factors and demographic factors on hazard perception of driver are multifaceted.