吉林大学学报(工学版) ›› 2025, Vol. 55 ›› Issue (12): 3875-3884.doi: 10.13229/j.cnki.jdxbgxb.20240537

• 交通运输工程·土木工程 • 上一篇    

联合博弈论与驾驶风格的混合交通流变道决策模型

郭昕刚1(),王嵩1,程超1(),范珍2   

  1. 1.长春工业大学 计算机科学与工程学院,长春 130012
    2.装甲兵学院士官学校,长春 130033
  • 收稿日期:2024-05-15 出版日期:2025-12-01 发布日期:2026-02-03
  • 通讯作者: 程超 E-mail:6889068@qq.com;125725673@qq.com
  • 作者简介:郭昕刚(1979-),男,教授,硕士.研究方向:人工智能.E-mail:6889068@qq.com
  • 基金资助:
    国家自然科学基金项目(62372063);长春市科技局重大专项项目(21GD05);吉林省科技厅重点攻关项目(20230508112RC)

Combined game theory and driving style hybrid traffic flow lane change decision model

Xin-gang GUO1(),Song WANG1,Chao CHENG1(),Zhen FAN2   

  1. 1.School of Computer Science and Engineering,Changchun University of Technology,Changchun 130012,China
    2.Armored Corps Academy Sub-lieutenant school,Changchun 130033,China
  • Received:2024-05-15 Online:2025-12-01 Published:2026-02-03
  • Contact: Chao CHENG E-mail:6889068@qq.com;125725673@qq.com

摘要:

针对人驾车(HV)与自动驾驶车辆(AV)混合交通流场景下车辆变道决策不准确的问题,提出了一种车辆变道决策模型。该模型基于博弈论,针对相邻车辆连续协同变道情形构建多重博弈函数,通过海萨尼变换消除通信不确定性的影响,并结合K-means++聚类区分车辆的驾驶风格,利用风险因子进一步调整博弈收益。运用SUMO仿真平台对变道模型进行验证,实验结果表明:在固定AV渗透率下,应用联合博弈论与驾驶风格的变道模型,车辆平均通过数量得到有效提高,且平均通过时间降低,同时测试中未发生事故,证明了变道模型的稳定性和安全性;在不同AV渗透率下,随着渗透率的升高,车辆的平均通过时间显著降低,表明AV能够有效利用车道。

关键词: 博弈论, 海萨尼变换, 驾驶风格, K-means++聚类, 变道决策

Abstract:

To address the issue of inaccuracy of vehicle lane change decision in the mixed traffic flow scenario of HV and AV, a vehicle lane change decision model was proposed. The model is based on game theory, multiple game functions were constructed for the continuous cooperative lane change situation of adjacent vehicles, the influence of communication uncertainty was eliminated by Harsanyi transform, and the driving style of vehicles was distinguished by K-means++ clustering, the game returns were further adjusted by risk factors. The lane change model was verified by using the SUMO simulation platform. The experimental results showed that under the fixed AV permeability, the average passing number of vehicles is effectively increased and the average passing time is reduced by applying the combined game theory and driving style lane change model, at the same time, no accidents occurred in the test, which proves the stability and safety of the lane change model. Under different AV permeability, the average passing time of vehicles decreases significantly with the increase of permeability, which indicates that the AV can effectively utilize the lane.

Key words: game theory, Harsanyi transform, driving style, K-means++ clustering, lane change decision

中图分类号: 

  • U491

图1

变道决策模型"

图2

典型变道场景"

图3

海萨尼变换"

表1

主车博弈收益矩阵"

后随车D策略主车A概率
等待变道

加速

减速

变道

不变

αP110+βQ110+τ11

αP110+βQ110+τ11

αP110+βQ110+τ11

αP110+βQ110+τ11

αP211+βQ211+αUV+UC+τ21

αP211+βQ221+αUV+UC+τ22

αP211+αUV+UC+τ23

αP211+βQ241+αUV+UC+τ24

n1

n2

n3

1-n1-n2-n3

概率m1-m

表2

后随车博弈收益矩阵"

后随车D策略主车A概率
等待变道

加速

减速

变道

不变

αP311+βQ311+UC+τ31

αP320+βQ320+UC+τ32

rΓ

αP340+βQ341+τ34

αP411+βQ311+αUV+UC+τ41

αP420+βQ320+αUV+UC+τ42

rΓ

αP440+βQ341+αUV+τ44

n1

n2

n3

1-n1-n2-n3

概率m1-m

表3

重复博弈收益矩阵"

J车策略J车收益D车收益

加速

减速

变道

不变

αP511+βQ511+αUV+UC+τ51

αP510+βQ520+αUV+UC+τ52

rΩ

αP540+βQ541+αUV+τ54

αP611+βQ611+αUV+UC+τ61

αP611+βQ621+αUV+UC+τ62

αP611+αUV+UC+τ63

αP611+βQ641+αUV+UC+τ64

表4

主成分特征值及贡献率"

成分特征值贡献率/%累计贡献率/%
13.78824.530 224.530 2
22.69817.468 341.998 5
32.42515.702 657.701 1
42.00712.992 970.694 0
51.56110.105 380.799 3
61.2398.023 888.823 1
70.5443.525 692.348 7
80.4442.876 595.225 2
90.2941.902 497.127 6
100.1681.090 198.217 7
110.0920.567 498.785 1
120.0640.415 799.200 8
130.0590.386 399.587 1
140.0340.213 699.800 7
150.0120.089 699.890 3
160.0110.073 299.963 5
170.0050.036 5100.000 0

表5

K-means++聚类结果"

聚类类别样本数量聚类中心
F1F2F3F4F5F6

1

2

3

4

16 984

19 673

9 405

5 304

0.303

-0.842

1.994

2.786

-0.381

0.523

1.264

2.388

-0.156

0.649

-1.002

0.951

0.073

-1.304

0.857

-1.516

0.059

-0.215

0.882

1.113

0.021

-0.035

-0.076

0.108

图4

不同时间窗分类"

表6

实验相关参数"

参数数值
车辆长度/m4.2
最高速度/(m·s-130
最大加速度/(m·s-23.3
最大减速度/(m·s-24.2

图5

仿真场景"

图6

不同车辆密度下对比"

表7

(拥堵,事故)次数"

车辆密度/(veh·h-1A组B组C组D组
1 000(0,0)(0,0)(0,0)(0,0)
1 500(0,0)(0,0)(0,0)(0,0)
2 000(1,0)(0,0)(1,0)(0,0)
2 500(1,1)(1,0)(1,1)(0,0)
3 000(2,1)(2,1)(2,1)(0,0)
3 500(2,2)(2,1)(3,2)(0,0)
4 000(3,2)(2,2)(4,4)(0,0)
4 500(3,2)(2,2)(6,5)(1,0)

表8

实验数据比较"

实验组平均通过时间/s通过数量/辆
1(96.78,90.35,82.56)(51,54,60)
2(102.54,98.72,90.21)(67,72,90)
3(112.43,102.57,97.35)(89,96,108)
4(139.56,128.33,120.49)(111,126,135)
5(151.28,141.71,129.65)(134,153,165)
6(176.12,173.49,156.53)(157,177,196)
7(191.53,180.52,170.02)(193,214,232)
8(212.36,203.66,181.18)(221,223,241)

图7

不同AV渗透率下平均通过时间"

[1] Gipps P G. A model for the structure of lane-changing decisions[J]. Transportation Research Part B: Methodological, 1986, 20(5): 403-414.
[2] Kita H. A game theoretic analysis of merging give-way interaction: a joint estimation model[J]. Transportation and Traffic Theory in the 21st Century, 2002(1): 503-518.
[3] Ji K, Han K. Optimal decision-making strategies for self-driving car inspired by game theory[C]∥International Conference on Ubiquitous and Future Networks. Piscataway, NJ: IEEE, 2021: 375-378.
[4] Peng H, Lv C, Huang C, et al. Cooperative decision making of connected automated vehicles at multi-lane merging zone: a coalitional game approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(4): 3829-3841.
[5] Peng H, Lv C, Xing X, et al. Human-like decision making for autonomous driving: a noncooperative game theoretic approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(4): 2076-2087.
[6] Peng H, Huang C, Hu Z X, et al. Decision making of connected automated vehicles at an unsignalized roundabout considering personalized driving behaviors[J]. IEEE Transactions on Vehicular Technology, 2021, 70(5): 4051-4064.
[7] 曲大义, 张可琨, 王韬, 等. 自动驾驶车辆换道决策行为分析及分子动力学建模[J]. 吉林大学学报: 工学版, 2022, 12(1): 1-13.
Qu Da-yi, Zhang Ke-kun, Wang Tao,et al. Analysis of lane change decision behavior and molecular dynamics modeling of autonomous vehicle[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 12(1): 1-13.
[8] 董俊一. 考虑驾驶风格的智能驾驶换道决策模型研究[D]. 长春: 吉林大学交通学院, 2022.
Dong Jun-yi. Research on lane change decision model of intelligent driving considering driving style[D]. Changchun: College of Transportation, Jilin University, 2022.
[9] 刘超, 丁菁. 一种基于合作博弈论的自动驾驶车道变更模型[J]. 计算机应用与软件, 2022, 39(1): 236-241.
Liu Chao, Ding Jing. An automated driving lane change model based on cooperative game theory[J]. Computers Applications and Software, 2022, 39(1): 236-241.
[10] 黑凯先, 曲大义, 郭海兵, 等. 网联混合车流车辆换道博弈行为及模型[J]. 科学技术与工程, 2021, 21(6): 2495-2501.
Kai-xian Hei, Qu Da-yi, Guo Hai-bing, et al. Game behavior and model of lane changing for networked mixed traffic flow vehicles[J]. Science, Technology and Engineering, 2021, 21(6): 2495-2501.
[11] Ali Y, Zheng Z D, Haque M, et al. A game theory-based approach for modelling mandatory lane-changing behavior in a connected environment[J]. Transportation Research Part C: Emerging Technologies, 2019, 106(12): 220-242.
[12] Elveve M, Ennana M. Gambit: Software tools for game theory[J]. Modern Healthcare, 2005, 41(5): 36-51.
[13] Krajewski R, Bock J, Kloeker L, et al. The highD dataset: a drone dataset of naturalistic vehicle trajectories on german highways for validation of highly automated driving systems[C]∥International Conference on Intelligent Transportation Systems, Maui, HI, USA, 2018: 2118-2125.
[14] 赵韩,刘浩,邱明明,等. 考虑车流密度影响的驾驶风格识别方法研究[J]. 汽车工程, 2020, 42(12): 1718-1727.
Zhao Han, Liu Hao, Qiu Ming-ming, et al. Research on driving style recognition method considering the influence of traffic flow density[J]. Automotive Engineering, 2020, 42(12): 1718-1727.
[15] 李经纬, 赵治国, 沈沛鸿, 等. 驾驶风格K-mean聚类与识别方法研究[J]. 汽车技术, 2018, 3(12): 8-12.
Li Jing-wei, Zhao Zhi-guo, Shen Pei-hong, et al. Research on K-means++ clustering and recognition of driving style[J]. Automotive Technology, 2018, 3(12): 8-12.
[16] Guo Y, Weng G R. K-means++ clustering-based active contour model for fast image segmentation[J]. Journal of Electronic Imaging, 2018, 27(6): 1-12.
[17] Wang R, Srdjan M L. Review of driving conditions prediction and driving style recognition based control algorithms for hybrid electric vehicles[C]∥Vehicle Power and Propulsion Conference, Chicago, IL, USA, 2011: 1-7.
[1] 张雅丽,付锐,袁伟,郭应时. 考虑能耗的进出站驾驶风格分类及识别模型[J]. 吉林大学学报(工学版), 2023, 53(7): 2029-2042.
[2] 肖雪,李克平,彭博,昌满玮. 基于决策-规划迭代框架的智驾车换道行为建模[J]. 吉林大学学报(工学版), 2023, 53(3): 746-757.
[3] 曲大义,黑凯先,郭海兵,贾彦峰,王韬. 车联网环境下车辆换道博弈行为及模型[J]. 吉林大学学报(工学版), 2022, 52(1): 101-109.
[4] 冯建鑫, 王挺峰, 郭劲. 测量丢失不确定性系统的迭代鲁棒滤波[J]. 吉林大学学报(工学版), 2014, 44(4): 1057-1061.
[5] 丛犁, 张海林, 刘毅, 赵力强, 张国鹏. 基于粒子群优化的协作网络资源分配的博弈策略[J]. 吉林大学学报(工学版), 2012, 42(01): 207-212.
[6] 林赐云1,2,龚勃文1,2,赵丁选1,3,杨兆升1,2. 基于博弈论的突发灾害下区域间交通信号协调控制技术[J]. 吉林大学学报(工学版), 2011, 41(05): 1257-1261.
[7] 于德新,杨兆升,王媛, 孙建平. 基于多智能体的城市道路交通控制[J]. 吉林大学学报(工学版), 2006, 36(01): 113-0118.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!