基于决策-规划迭代框架的智驾车换道行为建模

doi:10.13229/j.cnki.jdxbgxb20220560

摘要/Abstract

摘要：

行为决策和轨迹规划是智驾车完成驾驶任务的关键环节，着眼于二者之间的参数传递关系和新型混合交通流的动态特征，建立了智驾车换道模型。采用主从博弈思想描述智驾车换道决策过程，采用负指数函数量化博弈代价函数。将博弈决策结果作为轨迹规划模型输入之一，采用多项式分别描述车辆的横向及纵向运动，利用模拟退火算法寻找最优轨迹。设置了不同驾驶特性的环境车辆，并对决策规划过程进行了仿真验证，结果表明，决策规划模型能够快速生成安全、舒适的可行轨迹。

关键词: 交通运输系统工程, 博弈论, 轨迹规划, 模拟退火

Abstract:

Decision-making and motion planning are the crucial modules of autonomous vehicles. Focusing on the driving parameter relationship between decision making module and motion planning module， and the dynamic characteristics of new hybrid traffic flow， lane-changing model of autonomous vehicles was established. The decision-making process of lane-changing of autonomous vehicles was described by Stackelberg game theory， and the game cost function is quantified by negative exponential function. The decision results of the autonomous vehicle were taken as one of the inputs of the motion planning module. Polynomial was used to describe the lateral and longitudinal trajectories of the vehicle， and simulated annealing algorithm was used to find the optimal trajectory. The simulation results show that the proposed decision programming model can quickly generate safe， comfortable and feasible trajectory.

Key words: engineering of communications and transportation system, game theory, motion planning, simulated annealing

中图分类号:

U491.2

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.

Xue XIAO,Ke-ping LI,Bo PENG,Man-wei CHANG. Integrated lane⁃changing model of decision making and motion planning for autonomous vehicles[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(3): 746-757.

图/表 16

表1

博弈策略"

代价函数		跟随车辆
代价函数		加速 $a$	减速 $d$
领导车辆	换道 $c$	$(U a v, c, U h v, a)$	$(U a v, c, U h v, d)$
领导车辆	跟驰 $f$	$(U a v, f, U h v, a)$	$(U a v, c, U h v, d)$

表1

表2

轨迹规划代价函数"

含义	权重	代价分量
安全性	$w c$	$C c = 1 / ∑ 0 t D$
舒适性	$w J x$	$C J x = ∫ 0 t L ? A 2 t d t$
舒适性	$w J y$	$C J y = ∫ 0 t L ? O 2 t d t$
高效性	$w o$	$C o = ∫ 0 t L O 2 t d t$
	$w v$	$C v = L ˙ A t - v m a x 2$
	$w t$	$C t = t$

表2

图1

图2

表3

图3

图4

图5

图6

图7

表4

表5

表6

图8

图9

表7

参考文献 32

1	Hang P, Lv C, Huang C, et al. An integrated framework of decision making and motion planning for autonomous vehicles considering social behaviors[J]. IEEE Transactions on Vehicular Technology, 2020, 69(12): 14458-14469.
2	Lu Y, Xu X, Zhang X, et al. Hierarchical reinforcement learning for autonomous decision making and motion planning of intelligent vehicles[J]. IEEE Access, 2020, 8: 209776-209789.
3	Lee N, Choi W, Vernaza P, et al. Desire: distant future prediction in dynamic scenes with interacting agents[C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, America, 2017: 2156-2174.
4	Xu W, Pan J, Wei J, et al. Motion planning under uncertainty for on-road autonomous driving[C]∥2014 IEEE International Conference on Robotics and Automation (ICRA), Hong Kong, China, 2014: 2507-2512.
5	Schmerling E, Leung K, Vollprecht W, et al. Multimodal probabilistic model-based planning for human-robot interaction[C]∥2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, Australia, 2018: 1-9.
6	Gindele T, Brechtel S, Dillmann R. Learning driver behavior models from traffic observations for decision making and planning[J]. IEEE Intelligent Transportation Systems Magazine, 2015, 7(1): 69-79.
7	Schulz J, Hubmann C, Löchner J, et al. Interaction-aware probabilistic behavior prediction in urban environments[C]∥2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, 2018: 3999-4006.
8	Althoff M, Mergel A. Comparison of markov chain abstraction and monte carlo simulation for the safety assessment of autonomous cars[J]. IEEE Transaction on Intelligent Transportation System, 2011, 12(4): 1237-1247.
9	Kamrani M, Srinivasan A R, Chakraborty S, et al. Applying markov decision process to understand driving decisions using basic safety messages data[J]. Transportation Research Part C: Emerging Technologies, 2020, 115: 1-16.
10	Vallon C, Ercan Z, Carvalho A, et al. A machine learning approach for personalized autonomous lane change initiation and control[C]∥2017 IEEE Intelligent Vehicles Symposium (IV), Los Angeles, America, 2017: 1590-1595.
11	Lenz D, Diehl F, Le M T, et al. Deep neural networks for Markovian interactive scene prediction in highway scenarios[C]∥Intelligent Vehicles Symposium, Los Angeles, America, 2017: 685-692.
12	Xie D F, Fang Z, Jia B, et al. A data-driven lane-changing model based on deep learning[J]. Transportation Research Part C: Emerging Technologies, 2019, 106(2019): 41-60.
13	Morton J, Wheeler T A, Kochenderfer M J. Analysis of recurrent neural networks for probabilistic modeling of driver behavior[J]. IEEE Transaction on Intelligent Transportation System, 2017, 18(5): 1289-1298.
14	Yu H T, Tseng H E, Langari R, et al. A human-like game theory-based controller for automatic lane changing[J]. Transportation Research Part C: Emerging Technologies, 2018, 88(5): 140-158.
15	胡益恺, 庄瀚洋, 王春香, 等. 基于主从博弈的智能车汇流场景决策方法[J]. 上海交通大学学报, 2021, 55(8): 1027-1034.
	Hu Yi-kai, Zhuang Han-yang, Wang Chun-xiang, et al. Stackelberg-game-based intelligent vehicle decision method for merging scenarios[J]. Journal of Shanghai Jiaotong University, 2021, 55(8):1027-1034.
16	Kang K, Rakha H A. A repeated game freeway lane changing model[J]. Sensors, 2020, 20(6):1554-1588.
17	Hang P, Lv C, Huang C, et al. An integrated framework of decision making and motion planning for autonomous vehicles considering social behaviors[J]. IEEE Transactions on Vehicular Technology, 2020, 69(12): 14458-14469.
18	Katrakazas C, Quddus M, Chen W H, et al. Real-time motion planning methods for autonomous on-road driving: state-of-the-art and future research directions[J]. Transportation Research Part C: Emerging Technologies, 2015, 60: 416-442.
19	Gonzalez D, Pérez J, Milanes V, et al. A review of motion planning techniques for automated vehicles[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(4): 1135-1145.
20	Lim W, Lee S, Sunwoo M, et al. Hybrid trajectory planning for autonomous driving in on-road dynamic scenarios[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 22(1): 341-355.
21	Xu W D, Wei J Q, Dolan J M, et al. A real-time motion planner with trajectory optimization for autonomous vehicles[C]∥2012 IEEE International Conference on Robotics and Automation, St Paul, America, 2012: 2061-2067.
22	Ziegler J, Bender P, Dang T, et al. Trajectory planning for Bertha-A local continuous method[C]∥Intelligent Vehicles Symposium, Michigan, America, 2014: 450-457.
23	Jiang Y, Jin X, Xiong Y, et al. A dynamic motion planning framework for autonomous driving in urban environments[C]∥In Proceedings of the 2020 39th Chinese Control Conference, Shenyang, China, 2020: 5429-5435.
24	Liao X, Zhao X, Wu G, et al. A game theory based ramp merging strategy for connected and automated vehicles in the mixed traffic: a Unity-SUMO integrated platform[DB/OL]. [2022-01-27].
25	Zhou D, Ma Z, Sun J. Autonomous vehicles turning motion planning in multi-interaction area of mixed-flow intersections[J]. IEEE Transactions on Intelligent Vehicles, 2020, 5(2): 204-216.
26	Chu K, Lee M, Sunwoo M. Local path planning for off-road autonomous driving with avoidance of static obstacles[J]. IEEE Transactions on Intelligent Transportation Systems, 2012, 13(4): 1599-1616.
27	Werling M, Kammel S, Ziegler J, et al. Optimal trajectories for time-critical street scenarios using discretized terminal manifolds[J]. International Journal of Robotics Research, 2012, 31(3): 346-359.
28	陈华根, 吴健生, 王家林,等. 模拟退火算法机理研究[J]. 同济大学学报:自然科学版, 2004, 32(6): 802-805.
	Chen Hua-gen, Wu Jian-sheng, Wang Jia-lin, et al. Mechanism study of simulated annealing algorithm[J]. Journal of Tongji University (Natural Science), 2004, 32(6): 802-805.
29	Kirkpatrick S, Gelatt C D, Vecchi M P. Optimization by simulated annealing[J]. Science, 1983, 220(4598): 671-680.
30	唐俊. PSO算法原理及应用[J]. 计算机技术与发展, 2010, 20(2): 213-216.
	Tang Jun. Principle and application of PSO algorithm[J]. Computer Technology and Development, 2010, 20(2): 213-216.
31	Qiao F, Rahman R, Li Q, et al. Safe and environment-friendly forward collision warning messages in the advance warning area of a construction zone[J]. International Journal of Intelligent Transportation Systems Research, 2017, 15: 166-179.
32	Vishal M, Christos K, Constantinos A. Prediction of lane-changing maneuvers with automatic labeling and deep learning[J]. Transportation Research Record, 2020, 2674(7): 336-347.

相关文章 15

[1]	潘弘洋,刘昭,杨波,孙庚,刘衍珩. 基于新一代通信技术的无人机系统群体智能方法综述[J]. 吉林大学学报(工学版), 2023, 53(3): 629-642.
[2]	马敏,胡大伟,舒兰,马壮林. 城市轨道交通网络韧性评估及恢复策略[J]. 吉林大学学报(工学版), 2023, 53(2): 396-404.
[3]	王菁,万峰,董春娇,邵春福. 城市轨道交通站点吸引范围及强度建模[J]. 吉林大学学报(工学版), 2023, 53(2): 439-447.
[4]	方松,马健霄,李根,沈玲宏,徐楚博. 城市快速路右侧车道移动作业区行车风险分析[J]. 吉林大学学报(工学版), 2022, 52(8): 1786-1791.
[5]	宋现敏,杨舒天,刘明鑫,李志慧. 站点间公交行程时间波动特性及预测方法[J]. 吉林大学学报(工学版), 2022, 52(8): 1792-1799.
[6]	张玮,张树培,罗崇恩,张生,王国林. 智能汽车紧急工况避撞轨迹规划[J]. 吉林大学学报(工学版), 2022, 52(7): 1515-1523.
[7]	郑植,耿波,王福敏,董俊宏,魏思斯. 既有低等级混凝土护栏防护能力提升[J]. 吉林大学学报(工学版), 2022, 52(6): 1362-1374.
[8]	吴文静,战勇斌,杨丽丽,陈润超. 考虑安全间距的合流区可变限速协调控制方法[J]. 吉林大学学报(工学版), 2022, 52(6): 1315-1323.
[9]	徐洪峰,陈虹瑾,张栋,陆千惠,安娜,耿现彩. 面向网联汽车环境的单点全感应式信号配时技术[J]. 吉林大学学报(工学版), 2022, 52(6): 1324-1336.
[10]	盖松雪,曾小清,岳晓园,袁子豪. 基于用户-系统双层优化算法的车位引导模型[J]. 吉林大学学报(工学版), 2022, 52(6): 1344-1352.
[11]	李先通,全威,王华,孙鹏程,安鹏进,满永兴. 基于时空特征深度学习模型的路径行程时间预测[J]. 吉林大学学报(工学版), 2022, 52(3): 557-563.
[12]	冯天军,孙学路,黄家盛,田秀娟,宋现敏. 基于三种过街方式的两相位信号交叉口延误[J]. 吉林大学学报(工学版), 2022, 52(3): 550-556.
[13]	李兴华,冯飞宇,成诚,王洧,唐鹏程. 网约拼车服务选择偏好分析及建模[J]. 吉林大学学报(工学版), 2022, 52(3): 578-584.
[14]	尹超英,邵春福,黄兆国,王晓全,王晟由. 基于梯度提升决策树的多尺度建成环境对小汽车拥有的影响[J]. 吉林大学学报(工学版), 2022, 52(3): 572-577.
[15]	贾洪飞,邵子函,杨丽丽. 终点不确定条件下网约车合乘匹配模型及算法[J]. 吉林大学学报(工学版), 2022, 52(3): 564-571.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

参数	数值
车辆长度/m	4.2
车辆宽度/m	2
最小碰撞半径/m	3
加速度/（m·s^-2）	［-3，3］
车辆可行速度/（m·s^-1）	［0，17］
智驾车速度/（m·s^-1）	12.5
智驾车加速度/（m·s^-2）	0
智驾车驾驶特性参数	0.5，0.3，0.2
车1速度/（m·s^-1）	12.5
车1加速度/（m·s^-2）	0
车1驾驶特性参数	0.1，0.1，0.8
车2速度/（m·s^-1）	11
车2加速度/（m·s^-2）	0
车2驾驶特性参数	0.3，0.5，0.2
车3速度/（m·s^-1）	12.5
车3加速度/（m·s^-2）	0
车3驾驶特性参数	0.5， 0.3， 0.2

参数		取值
横向位置采样范围/m	左转	［1.8，5.4］
	直行	［-1.8，1.8］
	右转	［-5.4，-1.8］
横向位置采样间隔/m		0.1
规划时间采样范围/s		［4，5］
规划时间采样间隔/s		0.1
横向加速度采样范围/（m·s^-2）		［-3，3］
横向加速度采样间隔/（m·s^-2）		0.3
实际规划时间长度/s		0.1

参数	数值
初始温度	100
马尔科夫链长度	5
温度下降率	0.9
结束温度	3

含义	参数值
种群数量	5
最大迭代次数	20
惯性因子	0.95
加速常数	3

组别	起始温度	终止温度	降温比例0.9		降温比例0.95
组别	起始温度	终止温度	最优解	耗时/s	最优解	耗时/s
1	70	1	1.02	1.86	1.31	2.02
2	80	1	1.29	1.71	1.22	2.06
3	90	1	1.02	1.80	1.15	1.80
4	100	1	1.05	1.82	1.17	1.77
5	110	1	1.23	1.88	1.31	1.78
6	120	1	1.06	1.72	1.22	1.82
7	130	1	1.18	1.78	1.39	1.97