论文部分内容阅读
不同的道路平面线形几何设计对于驾驶人车道保持能力的需求是有差异的,驾驶人受疲劳程度影响也会呈现车道保持能力下降的趋势,当前的研究未综合考虑以上2个因素:线形和疲劳程度对驾驶横向表现的交互影响。邀请41位被试者分别开展550km的实车实验,获取车辆位置信息GPS以匹配道路线形类型,基于问卷调查方法获取驾驶过程疲劳等级。分析不同疲劳程度、不同平面线形类型以及弯道半径条件下的车道偏离标准差参数,构建了多元线性回归模型。数据分析结果表明,相同疲劳程度下驾驶人在圆曲线段驾驶的偏离值要超过直线段以及缓和曲线段;当弯道半径超过5 500m时,曲线段弯道半径越大,车道偏离差值越高。同时,考虑了线形影响的多元线性回归模型对疲劳程度的预测精度要高于未考虑线形因素的模型,进一步说明在针对驾驶疲劳行为表现开展研究时,有必要对道路设计参数加以考虑以提高疲劳辨识精度。
Different road geometries have different requirements for driver’s lane keeping ability. Driver’s fatigue tendency also shows the tendency of lane keeping ability to decrease. The current research does not consider the above two factors: linearity and fatigue Interaction of degree with driving horizontal performance. 41 subjects were invited to carry out a 550km real car experiment to obtain the vehicle position information GPS to match the road linear type, based on the questionnaire method to obtain the driving process fatigue level. The multi-linear regression model was constructed by analyzing the different degree of fatigue, different plane linear types and standard deviation parameters of lane departure under curve radius conditions. The results of the data analysis show that the deviation value of the driver driving in the circular curve segment exceeds the straight line segment and the easement curve segment under the same fatigue degree. When the curve radius exceeds 5 500m, the greater the curve radius of the curve segment is, high. At the same time, the multivariate linear regression model considering the influence of the linearity has a higher prediction accuracy on the degree of fatigue than the one without the linear factor, which further indicates that it is necessary to consider the road design parameters to improve the fatigue when conducting research on the behavior of driving fatigue Recognition accuracy.