考虑车辆历史状态数据的加速车道汇入策略研究

doi:10.16097/j.cnki.1009-6744.2023.04.011

交通运输系统工程与信息 ›› 2023, Vol. 23 ›› Issue (4): 102-110.DOI: 10.16097/j.cnki.1009-6744.2023.04.011

• 智能交通系统与信息技术 • 上一篇下一篇

考虑车辆历史状态数据的加速车道汇入策略研究

郭应时，谷梦路，王畅^*，苏彦奇，付锐，袁伟

长安大学，汽车学院，西安 710064

收稿日期:2023-03-28 修回日期:2023-05-06 接受日期:2023-05-08 出版日期:2023-08-25 发布日期:2023-08-21
作者简介:郭应时(1964- )，男，辽宁凌海人，教授，博士
基金资助:
国家自然科学基金(51908054)；国家重点研发计划 (2019YFB1600500)；中央高校基本科研业务费专项资金(CHD300102220202)

Merging Decisions from Acceleration Lanes Considering Historical Vehicle Operating State Data

GUO Ying-shi, GU Meng-lu, WANG Chang^*, SU Yan-qi, FU Rui, YUAN Wei

School of Automobile, Chang'an University, Xi'an 710064, China

Received:2023-03-28 Revised:2023-05-06 Accepted:2023-05-08 Online:2023-08-25 Published:2023-08-21
Supported by:
National Natural Science Foundation of China (51908054)；National Key Research and Development Program of China (2019YFB1600500)；Fundamental Research Funds for the Central Universities (CHD300102220202)

摘要/Abstract

摘要： 为研究汇入场景中车辆历史状态数据对高速公路加速车道汇入车辆汇入决策行为的影响，本文结合GentleBoost (Gentle adaptive Boosting)集成学习算法框架，提出考虑历史时间窗口的加速车道汇入决策模型。首先，使用高精度摄像头和毫米波雷达组成路侧数据采集平台，采集国内典型高速公路加速车道车辆汇入行为数据。其次，搭建汇入决策模型，基于汇入场景车辆当前时刻状态信息和历史状态信息，考虑剩余加速车道长度的影响，建立GentleBoost汇入决策模型。最后，通过SUMO (Simulation of Urban Mobility)仿真平台和MATLAB算法控制平台搭建智能网联高速公路加速车道汇入仿真测试环境，测试不同主线交通流密度下的汇入决策效果。研究结果表明，随着车辆历史状态数据时间窗口的增大，汇入决策模型的准确率先增大后趋于稳定。在考虑汇入场景车辆历史状态信息的时间窗口为1.7 s时，GentleBoost模型得到了最大的汇入决策识别准确率，其中识别“汇入”事件的准确率为98.9%，识别“不汇入”事件的准确率为97.4%。微观仿真结果表明，相比SUMO中的LC2013换道模型，考虑车辆历史状态信息的GentleBoost汇入决策模型获得了更高的汇入成功率和更大的通过平均速度。

关键词: 智能交通, 车辆历史状态数据, GentleBoost, 汇入决策模型, 加速车道, SUMO仿真验证

Abstract: To investigate the impact of the historical vehicle operating state data on merging decisions of the merging vehicle in a highway acceleration lane, a merging decision model for the merging vehicle considering the vehicle historical operating state data in a time window was proposed by combining with GentleBoost (Gentle adaptive Boosting) ensemble learning algorithm. First, a roadside data acquisition platform was designed using high�definition cameras and millimeter radar devices, and real traffic interaction data at a typical merging zone with an acceleration lane in China were collected. Second, a merging decision model based on a GentleBoost algorithm considering current and historical vehicle operating state data in the merging scenario, and the impact of the remaining distance in the acceleration lane was proposed. Finally, a microscopic simulation merging scenario was built based on SUMO (Simulation of urban mobility) and MATLAB platforms, to test the GentleBoost merging decision model in an intelligent connected environment under different traffic flow conditions. The results showed that, as the length of the time window increased, the overall recognition accuracy of the model first increased and finally tended to be stable. The merging decision model reached its best performance when the time window length was 1.7 s. And the recognition accuracy was 98.9% for "Merge" events and 97.4% for "Non-merge" events. The simulation results showed that, compared to the LC2013 model embedded in SUMO, the proposed GentleBoost model considering the historical data obtained a higher merging success rate and a larger average traveling speed in the merging control zone.

Key words: intelligent transportation, vehicle historical state data, GentleBoost, merging decision model, acceleration lane, SUMO simulation

中图分类号:

U495

郭应时, 谷梦路, 王畅, 苏彦奇, 付锐, 袁伟. 考虑车辆历史状态数据的加速车道汇入策略研究[J]. 交通运输系统工程与信息, 2023, 23(4): 102-110.

GUO Ying-shi, GU Meng-lu, WANG Chang, SU Yan-qi, FU Rui, YUAN Wei. Merging Decisions from Acceleration Lanes Considering Historical Vehicle Operating State Data[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(4): 102-110.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2023.04.011

http://www.tseit.org.cn/CN/Y2023/V23/I4/102

参考文献

[1] MORIDPOUR S, SARVI M, ROSE G. Lane changing models: A critical review[J]. Transportation Letters, 2010, 2(3): 157-173.

[2] 中华人民共和国公安部交通管理局. 中华人民共和国道路交通事故统计年报(2020 年度)[R]. 无锡: 公安部交通管理局, 2021. [Traffic Administration Bureau of The Ministry of Public Security of the Pepole's Republic of China. Annual report on road traffic accident statistics of the People's Republic of China[R]. Wuxi: Traffic Administration Bureau of the Ministry of Public Security 2021]

[3] LI L, GAN J, ZHOU K, et al. A novel lane-changing model of connected and automated vehicles: Using the safety potential field theory[J]. Physica A: Statistical Mechanics and its Applications, 2020, 559: 125039.

[4] WANG C, SUN Q, FU R, et al. Lane change warning threshold based on driver perception characteristics [J]. Accident Analysis & Prevention, 2018, 117: 164- 174.

[5] ALI Y, ZHENG Z, HAQUE MD M. Connectivity's impact on mandatory lane-changing behaviour: Evidences from a driving simulator study[J]. Transportation Research Part C: Emerging Technologies, 2018, 93: 292-309.

[6] 叶颖俊, 倪颖, 孙剑. 高密度瓶颈交通流主动-回应汇入行为定义与建模[J]. 中国公路学报, 2022, 35(8): 278-290. [YE Y J, NI Y, SUN J. Defining and modeling active-responsive merging behavior at high-density expressway on-ramp bottlenecks[J]. China Journal of Highway and Transport, 2022, 35(8): 278-290.]

[7] QI W, WANG W, SHEN B, et al. A modified post encroachment time model of urban road merging area based on lane-change characteristics[J]. IEEE Access, 2020, 8: 72835-72846.

[8] 赵建东, 贺晓宇, 余智鑫, 等. 多网联范围下的智能网联车换道决策组合模型研究[J]. 交通运输系统工程与信息, 2023, 23(1): 77-85. [ZHAO J D, HE X Y, YU Z X, et al. A combination model for connected and autonomous vehicles lane-changing decision-making under multi-connectivity range[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(1): 77-85.]

[9] 张洪加, 郭应时, 高松, 等. 网联与非网联环境下驾驶人换道意图识别研究[J/OL]. 中国公路学报,(2023-01-02) [2023-03-23]. http: //kns. cnki. net/ kcms/ detail/ 61. 1313. U. 20221230. 1243. 003. html. [ZHANG H J, GUO Y S, GAO S, et al. Research on driver lane- changing intention recognition in connected and non- connected environments[J]. China Journal of Highway and Transport, (2023- 01- 02) [2023- 03- 23]. http: //kns. cnki. net/ kcms/ detail/61.1313.U. 20221230. 1243. 003. html.]

[10] CHEEL O, SCHWARZ L, NAVAB N, et al. Situation assessment for planning lane changes: Combining recurrent models and prediction[C]. 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD: IEEE, 2018.

[11] 邬岚, 赵乐, 李根. 基于方差异质性随机参数模型的汇合行为分析[J]. 吉林大学学报(工学版), 2022(6): 1-7. [WU L, ZHAO L, LI G. Analysis of merging behavior based on random parameter model with heterogeneity in variances[J]. Journal of Jilin University (Engineering and Technology), 2022(6): 1-7.]

[12] CANTISANI G. Results of Micro-simulation model for exploring drivers' behavior on acceleration lanes[J]. European Transport/Trasporti Europei, 2020(77): 1-10.

[13] DOSHI A, TRIVEDI M. A comparative exploration of eye gaze and head motion cues for lane change intent prediction[C]. 2008 IEEE Intelligent Vehicles Symposium, Eindhoven, Netherlands: IEEE, 2008.

[14] SIMSEK M, KANTARCI B, ZHANG Y. Detecting fake mobile crowdsensing tasks: Ensemble methods under limited data[J]. IEEE Vehicular Technology Magazine, 2020, 15(3): 86-94.

[15] KRAUSS S, WAGNER P, GAWRON C. Metastable states in a microscopic model of traf fic flow[J]. Physical Review E, 1997, 55(5): 5597-5602.

考虑车辆历史状态数据的加速车道汇入策略研究

Merging Decisions from Acceleration Lanes Considering Historical Vehicle Operating State Data

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	李泰国, 张天策, 李超, 周星宏. 基于面部倒立摆模型与信息熵的驾驶员疲劳检测[J]. 交通运输系统工程与信息, 2023, 23(5): 24-32.
[2]	蒋贤才, 程国柱. 网联交通环境下交叉口进口车道动态配置方法[J]. 交通运输系统工程与信息, 2023, 23(5): 55-66.
[3]	李鹏程, 董宝田, 李思贤. 基于时间图注意力的交叉口交通状态识别及关联度研[J]. 交通运输系统工程与信息, 2023, 23(5): 67-74.
[4]	蔡浩, 李林峰, 李涵, 李新, 周腾. 基于极限学习机的短期交通流预测混合优化模型[J]. 交通运输系统工程与信息, 2023, 23(5): 75-82.
[5]	龙科军, 张仲根, 刘洋, 高志波. 城市道路应急车辆借道通行与信号协同优化模型[J]. 交通运输系统工程与信息, 2023, 23(5): 194-201.
[6]	方华珍, 刘立, 肖小凤, 顾青, 孟宇. 混合示教长短时记忆网络的车辆轨迹预测研究[J]. 交通运输系统工程与信息, 2023, 23(4): 80-87.
[7]	巫威眺, 周霄, 朱彦辰, 李鹏, 邹弘辉, 李余. 考虑乘客时空灵活性的需求响应客货联运动态调度[J]. 交通运输系统工程与信息, 2023, 23(4): 211-227.
[8]	郭煜东, 杨飞, 周涛, 姚振兴, 张楚良, 魏胤呈. 基于手机信令的城市机动化方式细分双层模型研究[J]. 交通运输系统工程与信息, 2023, 23(3): 101-109.
[9]	韩磊, 张轮, 郭为安. 混合交通流环境下基于改进强化学习的可变限速控制策略[J]. 交通运输系统工程与信息, 2023, 23(3): 110-122.
[10]	刘润坤, 于海洋, 程猛越, 任毅龙. 间歇性通信失效下网联自动驾驶队列控制策略[J]. 交通运输系统工程与信息, 2023, 23(2): 54-66.
[11]	张鹏, 顾云翔, 孙超, 李文权. 基于最小油耗的货车车速与信号配时协同优化模型[J]. 交通运输系统工程与信息, 2023, 23(2): 84-91.
[12]	路通, 余祖俊, 郭保青, 阮涛. 网状多尺度与双向通道注意力的铁路场景语义分割[J]. 交通运输系统工程与信息, 2023, 23(2): 233-241.
[13]	唐进君, 付强, 王骋程, 袁琛, 张璇, 徐长靖. 高速公路可变限速控制策略多目标优化[J]. 交通运输系统工程与信息, 2023, 23(2): 252-261.
[14]	张冰涛, 常文文, 李秀兰. 基于时空脑电特征与并行神经网络的疲劳驾驶检测[J]. 交通运输系统工程与信息, 2023, 23(2): 315-325.
[15]	赵建东, 贺晓宇, 余智鑫, 韩明敏. 多网联范围下的智能网联车换道决策组合模型研究[J]. 交通运输系统工程与信息, 2023, 23(1): 77-85.