智能网联车辆加速车道类人化汇入控制研究  

Human-like Merging Control of Intelligent Connected Vehicles on the Acceleration Lane

作　　者：谷梦路 葛振振 王畅[1,2] 苏彦奇 郭应时 GU Meng-lu;GE Zhen-zhen;WANG Chang;SU Yan-qi;GUO Ying-shi(School of Automobile,Chang'an University,Xi'an 710064,Shaanxi,China;Key Laboratory of Transportation Industry of Automotive Transportation Safety Enhancement Technology,Chang'an University,Xi'an 710064,Shaanxi,China)

机构地区：[1]长安大学汽车学院,陕西西安710064 [2]长安大学汽车运输安全保障技术交通行业重点实验室,陕西西安710064

出　　处：《中国公路学报》2024年第3期134-146,共13页China Journal of Highway and Transport

基　　金：国家自然科学基金项目(52102451,52002035);陕西省重点研发计划项目(2023-YBGY-035);中央高校基本科研业务费专项资金项目(CHD300102223501)。

摘　　要：为研究智能网联车辆(Intelligent Connected Vehicle, ICV)与高速公路主线传统人工驾驶车辆(Human-driven Vehicle, HDV)交互时通过高速公路加速车道的汇入控制算法,提出了一种融合随机森林(Random Forest, RF)算法和深度Q网络(Deep Q-network, DQN)算法的ICV汇入控制模型(DQN-RF)。首先,建立路侧数据采集平台,采集了中国G70高速公路加速车道汇入区域HDV的真实汇入过程数据。其次,考虑汇入环境车辆历史数据流和汇入车在加速车道的汇入紧迫度,建立了基于RF算法的类人化汇入决策模型。采用城市交通仿真(Simulation of Urban Mobility, SUMO)平台搭建了高速公路加速车道ICV汇入场景,并基于Python语言建立了ICV汇入控制深度强化学习测试脚本环境,建立了基于DQN的纵向加速度控制算法。最后,将RF汇入决策模型嵌入DQN纵向加速度控制算法中,实现了ICV汇入决策和纵向加速度控制的融合。将SUMO内置的LC2013换道模型与DQN模型融合为DQN-LC2013模型,作为基线模型与DQN-RF模型进行对比。研究结果表明:相较于未考虑类人化汇入决策的DQN-LC2013模型,考虑类人化汇入决策的DQN-RF模型获得了更高的奖励值;当加速度动作取值空间为[-1,2] m·s^(-2)时,DQN-RF和DQN-L2013控制下的ICV通过汇入控制区域的平均加速度分别为0.55、0.09 m·s^(-2),通过汇入控制区域的速度分别为21.4、19.7 m·s^(-1);DQN-RF模型控制下ICV无停车等待现象,DQN-LC2013控制下ICV在100次汇入过程中出现了7次停车等待。提出的DQN-RF汇入控制模型可以实现类人化的驾驶决策,提高ICV通过汇入控制区域的效率和汇入成功率,可用于ICV通过高速公路加速车道汇入区域的汇入决策和纵向加速度控制。To develop a merging control algorithm for intelligent connected vehicles(ICVs)on freeway acceleration lanes interacting with human-driven vehicles(HDVs)on the mainline,we propose a merging control model(DQN-RF).This model integrates the deep Q-network(DQN)algorithm and the random forest(RF)algorithm.First,a roadside data acquisition platform was established to collect the naturalistic merging behavior data of HDVs at a typical merging zone with an acceleration lane on the G70freeway in China.Second,a human-like merging decision model using RF was built using historical merging environmental contextual data and the merging urgency of the merging vehicle on the acceleration lane as input.We constructed a simulated merging scenario featuring an acceleration lane on the freeway using the simulation of urban mobility(SUMO)platform.Utilizing the Python language,we developed a testing script environment for the deep reinforcement learning algorithm.Additionally,we introduced a longitudinal acceleration control algorithm based on DQN.Finally,the DQN-RF merging control model,which embedded the RF merging decision algorithm into the DQN longitudinal acceleration control algorithm,was established to embrace merging decision control and longitudinal acceleration control in a comprehensive framework.The default lane-changing control algorithm in SUMO,known as"LC2013,"was also combined with the proposed DQN algorithm to serve as a baseline model.The simulation results show that,with the same acceleration action value space[-1,2]m·s^(-2),compared to the DQN-LC2013model,the DQN-RF model receives a higher total reward value.The average accelerations of the ICV for the DQN-RF and DQN-LC2013models are 0.55and 0.09m·s^(-2),respectively.Furthermore,the average speeds are 21.4and 19.7m·s^(-1),respectively.There are no stop-and-wait phenomena observed when the DQN-RF model is adopted,while there are seven stop-and-wait events in 100 turns of simulation when the DQN-LC2013 model is adopted.The proposed DQN-RF merging control model c

关 键 词：交通工程 汇入控制模型 DQN-RF 智能网联汇入车辆 加速车道 SUMO仿真 

分 类 号：U491.2[交通运输工程—交通运输规划与管理]