基于跟驰对的CO2排放特性的生态车辆跟驰策略

doi:10.3969/j.issn.1674-8484.2025.04.008

摘要/Abstract

摘要：

提出了一种在智能网联汽车环境下的混合交通流的生态车辆跟驰（ECF）策略，探索其跟驰行为的CO₂排放。基于车辆轨迹数据，提取多测度的跟驰行为特征参数。建立了极端梯度提升（XGBoost）模型，利用加性解释（SHAP）算法计算并分析了跟驰行为特征参数对跟驰过程中CO₂排放量的影响规律。标定了人工驾驶车辆的智能驾驶员模型，并基于城市交通仿真（SUMO）平台以及自动驾驶车辆（CAV）的自适应巡航控制（ACC）、协同式自适应巡航控制（CACC）模型；在11个混合交通流仿真场景下，分析ECF策略的CO₂减排有效性。结果表明：当CACC车辆占比50%以上时，CACC-CACC跟驰对的CO₂瞬时质量排放量减少比例超过60%。从而，本文ECF策略能够降低车辆在混合交通流场景下跟驰过程中CAV和CACC-CACC跟驰对的CO₂排放。

关键词: 自动驾驶车辆（CAV）, 混合交通流, 生态车辆跟驰（ECF）策略, 跟驰行为, 城市交通仿真（SUMO）平台

Abstract:

An eco-car-following (ECF) strategies was explored with the CO₂ emissions of car-following behavior in mixed traffic flow under the environment of intelligent connected vehicles. The vehicle trajectory data was used to extract multi-dimensional car-following behavior feature parameters. An eXtreme Gradient Boosting (XGBoost) model was established with calculating and analyzing the effects of car-following behavior feature parameters on CO₂ emissions during the car-following process by using the Shapley Additive exPlanations (SHAP) algorithm. The intelligent driver model of human-driven vehicles was calibrated. The Simulation of Ur-ban MObility (SUMO) platform was using to simulate 11 mixed traffic scenarios. The Adaptive Cruise Control (ACC) and the Cooperative Adaptive Cruise Control (CACC) models were employed for Connected and Automated Vehicles (CAVs). The results show that the instantaneous mass CO₂ emissions of CACC-CACC vehicle pairs de-crease by more than 60% when the proportion of CACC vehicles exceeds 50%. There-fore, the strategy reduces CO₂ emissions for CAVs and CACC-CACC car-following pairs in mixed traffic flow scenarios.

Key words: connected automated vehicles (CAVs), mixed traffic flow, eco-car-following (ECF) strategy, car-following behavior, simulation of urban mobility (SUMO) platform

中图分类号:

U491

于谦, 郭圆圆, 杨鸣鹏, 张玉婷. 基于跟驰对的CO₂排放特性的生态车辆跟驰策略[J]. 汽车安全与节能学报, 2025, 16(4): 577-586.

YU Qian, GUO Yuanyuan, YANG Mingpeng, ZHANG Yuting. Eco-car-following strategy based on the CO₂ emission characteristics of car-following pairs[J]. Journal of Automotive Safety and Energy, 2025, 16(4): 577-586.

图/表 10

参考文献 19

[1]	李国法, 陈耀昱, 吕辰, 等. 智能汽车决策中的驾驶行为语义解析关键技术[J]. 汽车安全与节能学报, 2019, 10(4): 391-412.
	LI Guofa, CHEN Yaoyu, LÜ Chen, et al. Key techniques of semantic analysis of driving behavior in decision making of autonomous vehicles[J]. J Autom Safe Energ, 2019, 10(4): 391-412. (in Chinese)
[2]	张铎, 饶红玉, 刘佳琦, 等. 基于自然驾驶数据的驾驶人跟驰行为内在异质性预测与建模[J]. 交通运输系统工程与信息, 2023, 23(5): 33-44.
	ZHANG Duo, RAO Hongyu, LIU Jiaqi, et al. Intra-driver heterogeneity prediction and modeling based on naturalistic driving experiment[J]. J Transp Syst Eng Info Techn, 2023, 23(5): 33-44. (in Chinese)
[3]	冉斌, 谭华春, 张健, 等. 智能网联交通技术发展现状及趋势[J]. 汽车安全与节能学报, 2018, 9(2): 119-130.
	RAN Bin, TAN Huachun, ZHANG Jian, et al. Development status and trend of connected automated vehicle highway system[J]. J Autom Safe Energ, 2018, 9(2): 119-130. (in Chinese)
[4]	杨秀建, 黄菁菁, 王曦. 考虑CACC汽车队列动态特性的混合交通流建模与仿真[J]. 系统仿真学报, 2024, 24(175): 1-13.
	YANG Xiujian, HUANG Jingjing, WANG Xi. Modeling and simulation of hybrid traffic flow considering the inherent dynamics of CACC vehicular platoons[J]. J Syst Simul, 2024, 24(175): 1-13. (in Chinese)
[5]	LI Renjie, QIN Yanyan. Car-following strategy involving stabilizing traffic flow with connected automated vehicles to reduce particulate matter (PM) emissions in rainy weather[J]. Sustain, 2024, 16(5): 2045.
[6]	HUANG Yan, YAN Xuedong, LI Xiaomeng, et al. Evaluating the safe and eco-driving performances of car-following behaviors in a vehicle platoon under foggy conditions[J]. J Transp Saf Secur, 2024, 16(2): 204-229.
[7]	QIN Yanyan, XIAO Tengfei, HE Zhengbing. Emissions-reduction strategy for connected autonomous vehicles on mixed traffic freeways[J]. Phys A, 2024, 653: No 130113.
[8]	HU Xiaosong, ZHANG Xiaoqian, TANG Xiaolin, et al. Model predictive control of hybrid electric vehicles for fuel economy, emission reductions, and inter-vehicle safety in car-following scenarios[J]. Energy, 2020, 196: Paper No 117101.
[9]	李传耀, 张帆, 王涛, 等. 基于深度强化学习的道路交叉口生态驾驶策略研究[J]. 交通运输系统工程与信息, 2024, 24(01): 81-92.
	LI Chuanyao, ZHANG Fan, WANG Tao, et al. Signalized Intersection Eco-driving Strategy Based on Deep Reinforcement Learning[J]. J Transp Syst Eng Info Techn, 2024, 24(01): 81-92. (in Chinese)
[10]	DONG Haoxuan, ZHUANG Weichao, CHEN Boli, et al. A comparative study of energy-efficient driving strategy for connected internal combustion engine and electric vehicles at signalized intersections[J]. Appl Energy, 2022, 310: No 118524.
[11]	秦严严, 王昊, 冉斌. CACC车辆跟驰建模及混合交通流分析[J]. 交通运输系统工程与信息, 2018, 18(2): 60-65.
	QIN Yanyan, WANG Hao, RAN Bin. Car-following modeling for CACC vehicles and mixed traffic flow analysis[J]. J Transp Syst Eng Info Techn, 2018, 18(2): 60-65. (in Chinese)
[12]	Federal Highway Administration. US highway 101 dataset[Z/OL]. (2007-01-01). https://www.fhwa.dot.gov.
[13]	Chong L S, Abbas M M, Flintsch A M, et al. A rule-based neural network approach to model driver naturalistic behavior in traffic[J]. Transp Res Pt C-Emerg Technol, 2013, 32: 207-223.
[14]	XU Wei, LI Nan, LIANG Wenli, et al. Dissipation characteristics of vehicle queue in V2X environment based on improved car-following model[J]. Math Probl Eng, 2022, 2022: 1-17.
[15]	王雪松, 朱美新. 基于自然驾驶数据的中国驾驶人城市快速路跟驰模型标定与验证[J]. 中国公路学报, 2018, 31(9): 129-137.
	WANG Xuesong, ZHU Meixin. Calibrating and validating car-following models on urban expressways for chinese drivers using naturalistic driving data[J]. China J Highw Transp, 2018, 31(9): 129-137. (in Chinese)
[16]	Lundberg S, Lee S I. Welcome to the SHAP documentation[Z/OL]. (2025-03-05). https://shap.readthedocs.io/en/latest/.
[17]	Treiber M, Hennecke A, Helbing D. Congested traffic states in empirical observations and microscopic simulations[J]. Phys Rev E, 2000, 62(2): 1805-1824.
[18]	曹金亮, 史忠科, 房雅灵. 基于城市主干路交通流数据的跟驰模型标定[J]. 交通信息与安全, 2014, 32(6): 82-88.
	CAO Jinliang, SHI Zhongke, FANG Yaling. Calibrating car-following models based on urban arterial video data[J]. J Transp Info Saf, 2014, 32(6): 82-88. (in Chinese)
[19]	秦严严, 杨晓庆, 王昊. 智能网联混合交通流CO₂排放影响及改善方法[J]. 吉林大学学报:工学版, 2023, 53(1): 150-158.
	QIN Yanyan, YANG Xiaoqing, WANG Hao. Impacts of CO₂ emissions and improving method for connected and automated mixed traffic flow[J]. J Jilin Univ: Eng Techn Ed, 2023, 53(1): 150-158. (in Chinese)

标定参数	默认值	标定范围	标定结果
v₀(t) / (m·s^-1)	不设置上限	[15, 40]	17.61
T₀(t) / s	0.25	[0.1, 5]	0.10
S_jam / m	2.5	[0.1, 10]	0.10
a_max / (m·s^-2)	2.6	[0.1, 5]	0.39
a_comf / (m·s^-2)	4.5	[0.1, 5]	2.84
β	4	20	20

标定参数	默认值	标定范围	标定结果
v₀(t) / (m·s^-1)	不设置上限	[15, 40]	17.61
T₀(t) / s	0.25	[0.1, 5]	0.10
S_jam / m	2.5	[0.1, 10]	0.10
a_max / (m·s^-2)	2.6	[0.1, 5]	0.39
a_comf / (m·s^-2)	4.5	[0.1, 5]	2.84
β	4	20	20

参数	跟驰模型
参数	HV	CACC	ACC
颜色	红色	绿色	蓝色
车辆长度 / m	5	5	5
车辆加速能力 / (m·s^-2)	0.39	2.6	2.6
车辆减速能力 / (m·s^-2)	2.84	4.5	4.5
车辆最大速度 / (m·s^-1)	17.61	34	34
车辆与前车距离 / m	0.10	2.5	2.5
加速度指数	20	-	-
车头时距 / s	0.10	1/2*	1/2*
车辆出发速度 / (m·s^-1)	随机	随机	随机
驾驶员完美程度	0.5	0.5/0*	0.5/0*

参数	跟驰模型
参数	HV	CACC	ACC
颜色	红色	绿色	蓝色
车辆长度 / m	5	5	5
车辆加速能力 / (m·s^-2)	0.39	2.6	2.6
车辆减速能力 / (m·s^-2)	2.84	4.5	4.5
车辆最大速度 / (m·s^-1)	17.61	34	34
车辆与前车距离 / m	0.10	2.5	2.5
加速度指数	20	-	-
车头时距 / s	0.10	1/2*	1/2*
车辆出发速度 / (m·s^-1)	随机	随机	随机
驾驶员完美程度	0.5	0.5/0*	0.5/0*

场景	CAV渗透率 / %	交通量/ (辆·h^-1)	1 h 交通量 / 辆			CACC/ACC 车辆比例
场景	CAV渗透率 / %	交通量/ (辆·h^-1)	HV车辆	CACC车辆	ACC车辆	CACC/ACC 车辆比例
1*	50	1 600	800	400	400	1∶1
2	25	1 600	1 200	200	200	1∶1
3	75	1 600	400	600	600	1∶1
4	100	1 600	0	800	800	1∶1
5	50	800	400	200	200	1∶1
6	50	2 400	1 200	600	600	1∶1
7	50	3 200	1 600	800	800	1∶1
8	50	1 600	800	0	800	0∶800
9	50	1 600	800	200	600	1∶3
10	50	1 600	800	600	200	3∶1
11	50	1 600	800	800	0	800∶0

基于跟驰对的CO₂排放特性的生态车辆跟驰策略

Eco-car-following strategy based on the CO₂ emission characteristics of car-following pairs

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 19

相关文章 15

编辑推荐

Metrics

本文评价

期刊信息

在线期刊

作者中心

审稿中心

联系我们

CAV渗透率 %	CO₂瞬时质量排放 / (g·s^-1)		减少比例 %
CAV渗透率 %	不使用ECF策略	使用ECF策略	减少比例 %
50	21.30	6.72	68.48
75	19.48	6.89	64.63
100	17.87	6.95	61.11
交通量 / (辆 · h^-1)
1 600	21.30	6.72	68.48
2 400	7.71	6.72	12.92
3 200	7.93	6.63	16.34
CACC比例 / %
50	21.30	6.72	68.48
75	19.00	6.94	63.50
100	17.51	6.94	60.40

[1]	韩雨, 陈志轩, 王翊萱, 李春杰, 雷伟, 焦彦利, 刘攀. 基于深度强化学习的入口匝道流量调控方法[J]. 汽车安全与节能学报, 2025, 16(4): 587-597.
[2]	杨阳, 王奕淳, 张义, 于谦, 张涛. 基于MOVES模型的太原市道路机动车氨气排放特征[J]. 汽车安全与节能学报, 2025, 16(3): 425-433.
[3]	虞安军, 励英迪, 杨哲懿, 付崇宇, 童蔚苹, 余佳, 刘云海, 刘志远. 基于多维注意力机制的高速公路交通流量预测方法[J]. 汽车安全与节能学报, 2025, 16(3): 463-469.
[4]	徐军莉. 基于功能性脑网络和图卷积网络的驾驶疲劳检测[J]. 汽车安全与节能学报, 2025, 16(2): 226-233.
[5]	杨澜, 赵祥模, 王润民, 王振, 房山, 瞿广跃. 自动驾驶认知能力测试评价研究综述[J]. 汽车安全与节能学报, 2025, 16(1): 1-15.
[6]	吴庆富, 袁满荣, 郝帅杰, 刘嘉伟, 牛世峰, 刘金峰. 紧急状况下驾驶人应激反应的特性指标[J]. 汽车安全与节能学报, 2025, 16(1): 43-49.
[7]	刘擎超, 王瑞海, 蔡英凤, 王海, 陈龙. 基于CatBoost和SHAP的高级别自动驾驶车辆非预期停车冲突风险预测[J]. 汽车安全与节能学报, 2025, 16(1): 170-180.
[8]	李昕光, 孙崇效, 曲大义, 于文昌, 胡含. 考虑交叉口启停波的智能网联车混行生态驾驶策略[J]. 汽车安全与节能学报, 2024, 15(6): 895-904.
[9]	蔡田茂, 孔伟伟, 罗禹贡, 石佳, 姬鹏霄, 李聪民. 基于MADDPG算法的匝道合流区多车协同控制[J]. 汽车安全与节能学报, 2024, 15(6): 923-933.
[10]	刘洋, 占佳豪, 李深, 李小鹏, 陈峻. 自动驾驶技术的未来：单车智能和智能车路协同[J]. 汽车安全与节能学报, 2024, 15(5): 611-633.
[11]	吴桐, 黄凯, 刘志远, 蒋玮. 交通网络和电力网络融合承载力的研究综述[J]. 汽车安全与节能学报, 2024, 15(5): 634-649.
[12]	姚荣涵, 徐文韬, 林子敬, 王立冰. 公交-合乘车道优化设计的多目标双层规划模型[J]. 汽车安全与节能学报, 2024, 15(5): 711-722.
[13]	张永升, 李逸椆, 王亮, 徐志刚. 智能网联环境下车载终端的发展现状与挑战[J]. 汽车安全与节能学报, 2024, 15(3): 295-308.
[14]	陈春, 王晨宇, 张道文. 基于时空结合的道路交通事故的黑点判别方法[J]. 汽车安全与节能学报, 2024, 15(3): 321-328.
[15]	王金萍, 冯海霞, 赵欢欢, 韩国华, 霍苗苗, 施庆利, 宁二伟. 一种新的机动车排放清单的时间分配方法[J]. 汽车安全与节能学报, 2024, 15(3): 387-394.