摘要
为了提高自适应巡航控制跟驰模型的效率和安全性,考虑不同驾驶员驾驶行为的差异性,根据驾驶员以往的驾驶行为特性,利用k-均值聚类算法对驾驶风格进行判别和分类,作为优化自适应巡航控制跟驰行为的依据;提出智能网联环境中自适应巡航控制跟驰优化方法,基于对不同驾驶风格车辆的动力学分析,引入驾驶风格修正系数、安全冗余修正系数、响应延迟时间,针对不同前车驾驶风格,建立改进的自适应巡航控制跟驰模型,并对所建立的模型、原自适应巡航控制跟驰模型及对比模型进行仿真分析。结果表明,相比原自适应巡航控制跟驰模型和对比模型,所建立模型的加速度曲线和车头间距曲线均更平缓,可以有效提高跟驰效率,同时具有更高的安全性。
To improve efficiency and safety of adaptive cruise control car-following models,considering differences in driving behaviors of different drivers and according to previous driving behavior characteristics of the drivers,driving styles was discriminated and classified by using k-means clustering algorithm,which were used as a basis for optimizing the adaptive cruise control car-following behavior.Proposing an adaptive cruise control car-following optimization method under intelligent network connection environment,based on dynamic analysis of vehicles with different driving styles,and introducing driving style correction coefficient,safety redundancy correction coefficient as well as response delay time,an improved adaptive cruise control car-following model was established according to different driving styles of the leading car.The established model,the original adaptive cruise control car-following model,and the comparative model were simulated and analyzed.The results show that acceleration curves and space headway curves of the established model are gentler than those of the original ad aptive cruise control car-following model and the comparative model.The efficiency of car-following is effectively improved and the safety is higher.
作者
胡春燕
曲大义
赵梓旭
宋慧
王韬
HU Chunyan;QU Dayi;ZHAO Zixu;SONG Hui;WANG Tao(School of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao 266525,Shandong,China;Shandong Branch of Nanjing Institute of City and Transport Planning Co.,Ltd.,Qingdao 266000,Shandong,China)
出处
《济南大学学报（自然科学版）》
CAS
北大核心
2023年第3期331-338,共8页
Journal of University of Jinan(Science and Technology)
基金
国家自然科学基金项目(62003182)
山东省重点研发计划项目(2019GGX101038)。
关键词
交通工程
跟驰模型
模型仿真
自适应巡航控制
驾驶风格
K-均值聚类算法
traffic engineering
car-following model
model simulation
adaptive cruise control
driving style
k-means clustering algorithm
