网联异质车群的自主换道行为及动力学建模  

Autonomous Lane Change Behavior and Dynamics Modeling of Networked Heterogeneous Vehicle Clusters

作　　者：杨子奕 曲大义[1] 王韬 陈意成 邵德栋 YANG Zi-yi;QU Da-yi;WANG Tao;CHEN Yi-cheng;SHAO De-dong(College of Mechanical and Automotive Engineering,Qingdao University of Technology,Qingdao 266520,china)

机构地区：[1]青岛理工大学交通运输工程系,青岛266520

出　　处：《科学技术与工程》2024年第15期6473-6484,共12页Science Technology and Engineering

基　　金：国家自然科学基金(52272311)。

摘　　要：道路上行驶的网联异质车流,存在着网联自主驾驶汽车(connected autonomous vehicles,CAVs)和人工驾驶汽车(human-driven vehicles,HVs)的复杂交互耦合关系。传统的交通优化模型着重于均匀化网联车辆的运行速度,缺乏对网联车辆自主换道和主动礼让行为的考虑。为引导CAV自主变道决策和让行行为,挖掘其自主行为的动力学特性,运用系统动力学建立车车交互行为模型,判断CAVs所处的空间阈值,控制瓶颈区CAVs的目标巡航速度,调节瓶颈区道路车辆密度,提高异质车流通行效率。采用MATLAB数值仿真,验证分析面向CAV自主决策行为的动力学模型,结果表明:相较于传统的交通优化模型,加入符合换道条件和让行条件下的CAVs自主行为因素后,交通瓶颈区车辆平均延误以及瓶颈区排队长度均减少约20%,其中停车延误可减少约40%。网联异质车流中车车交互行为特性及自主决策行为动力学模型可为提高异质车流在瓶颈区的高效运行提供理论基础。In order to guide CAVs autonomous lane-changing decisions and yielding behaviors,and to explore the dynamic characteristics of their autonomous behaviors,a system dynamics model of vehicle-to-vehicle interactions was established.The model determines the spatial threshold for CAVs,controls the target cruising speed of CAVs in bottleneck areas,adjusts the traffic density of road vehicles in bottleneck areas,and improves the efficiency of heterogeneous traffic flow.MATLAB numerical simulations were conducted to validate the dynamic model for CAVs autonomous decision-making behaviors.The results show that,compared to traditional traffic optimization models,incorporating autonomous behaviors factors for CAVs under lane-changing and yielding conditions reduces the average delay of vehicles in bottleneck areas and the queue length by approximately 20%,with parking delays reduced by about 40%.It is concluded that the characteristics of vehicle-to-vehicle interactions and the dynamic model of autonomous decision-making behaviors in connected heterogeneous traffic flow can provide a theoretical basis for improving the efficient operation of heterogeneous traffic flow in bottleneck areas.

关 键 词：网联异质车群 智能网联汽车 自主决策行为 系统动力学 

分 类 号：U491.2[交通运输工程—交通运输规划与管理]