基于多指标耦合的自动驾驶汽车实时轨迹规划算法

doi:10.3969/j.issn.1674-8484.2023.04.010

摘要/Abstract

摘要：

城市交通环境对自动驾驶汽车的安全性、舒适性、合规性、通行效率等指标提出了更高的要求。该文提出了一种时空耦合的实时轨迹规划方法，在其评价函数中引入上述多种量化指标。该算法在一组驾驶走廊内生成轨迹，根据结合横纵向舒适性、安全性和效用性的多指标评价函数选择最优轨迹，并在模拟环岛中进行了测试。结果表明：当舒适性优先时，加速度曲线较为平滑，最大值为0.66 m/s²; 当安全性优先时，产生的轨迹平均偏离0.19 m，避免与附近障碍物发生碰撞; 当效用性优先时，生成的轨迹最大速度为9.97 m/s，并表现出较积极的加速度场。该算法可使自动驾驶汽车在保持横纵向加速度的同时安全避开动态障碍物，且每次迭代平均耗费47.45 ms，满足实时性的要求。

关键词: 自动驾驶, 轨迹规划, 多指标耦合, 评价函数, 实时性

Abstract:

Urban traffic environment puts forward higher requirements for the safety, comfort, compliance, traffic efficiency and other indicators of autonomous vehicles. A spatiotemporal coupled real-time trajectory planning method was proposed, and the above quantitative indexes were introduced into its evaluation function. The algorithm generated trajectories in a set of driving corridors, selected the optimal trajectories according to a multi-index evaluation function combining horizontal and longitudinal comfort, safety and utility, and tested them in a simulated roundabout. The results show that the acceleration curve is smooth and the maximum value is 0.66 m/s² when comfort is the priority; The generated trajectory is deviated by 0.19 m on average to avoid collision with nearby obstacles when safety comes first; The generated trajectory has a maximum velocity of 9.97 m/s and exhibits a more positive acceleration profile when utility takes precedence. The algorithm can safely avoid dynamic obstacles while maintaining transverse and longitudinal acceleration, and the average cost of each iteration is 47.45 ms, which meets the real-time requirements.

Key words: autonomous driving, trajectory planning, multi-index coupling, evaluation function, real-time

中图分类号:

U461.91

黄鹏程, 裴晓飞, 周洪龙, 陈词. 基于多指标耦合的自动驾驶汽车实时轨迹规划算法[J]. 汽车安全与节能学报, 2023, 14(4): 480-487.

HUANG Pengcheng, PEI Xiaofei, ZHOU Honglong, CHEN Ci. Trajectory planning algorithm of autonomous vehicle based on multi-index coupling[J]. Journal of Automotive Safety and Energy, 2023, 14(4): 480-487.

图/表 15

参考文献 20

[1]	李克强, 戴一凡, 李升波, 等. 智能网联汽车(ICV)技术的发展现状及趋势[J]. 汽车安全与节能学报, 2017, 8(1): 1-14.
	LI Keqiang, DAI Yifan, LI Shengbo, et al. State-of-the-art and technical trends of intelligent and connected vehicles[J]. J Autom Safe Energ, 2017, 8(1): 1-14. (in Chinese)
[2]	Claussmann L, Revilloud M, Gruyer D, et al. A review of motion planning for highway autonomous driving[J]. IEEE Trans Intel Transp Syst, 2019, 21(5), 1826-1848. doi: 10.1109/TITS.6979 URL
[3]	Yurtsever E, Lambert J, Carballo A, et al. A survey of autonomous driving: common practices and emerging technologies[J]. IEEE Access, 2020, 8: 58443-58469. doi: 10.1109/Access.6287639 URL
[4]	余卓平, 李奕姗, 熊璐. 无人车运动规划算法综述[J]. 同济大学学报(自然科学版), 2017, 45(8): 1150-1159.
	YU Zhuoping, LI Yishan, XIONG Lu. A review of motion planning algorithms for autonomous vehicles[J]. J Tongji Univ (Nat Sci Edit), 2017, 45(8): 1150-1159. (in Chinese)
[5]	ZHENG Ling, ZENG Pengyun, YANG Wei, et al. Bezier curve-based trajectory planning for autonomous vehicles with collision avoidance[J]. IET Intel Transport Syst, 2020, 14(13), 1882-1891.
[6]	MENG Fei, SHEN Xuyu, Karimi H R. Emerging methodologies in stability and optimization problems of learning-based nonlinear model predictive control: a survey[J]. Int J Circu Theor Appl, 2022, 50(11), 4146-4170.
[7]	ZHANG Ting, FU Mengyin, SONG Wenjie, et al. Trajectory planning based on spatio-temporal map with collision avoidance guaranteed by safety strip[J]. IEEE Trans Intel Transp Syst, 2022, 23(2): 1030-1043. doi: 10.1109/TITS.2020.3019514 URL
[8]	XIN Long, KONG Yiting, LI Shengbo, et al. Enable faster and smoother spatio-temporal trajectory planning for autonomous vehicles in constrained dynamic environment[J]. Proceed Instit Mech Engi Part D: J Autom Engi, 2021, 235(4): 1101-1112.
[9]	Karaman S, Frazzoli E. Sampling-based algorithms for optimal motion planning[J]. Int’l J Robot Res, 2011, 30(7): 846-894.
[10]	Rocamora B M, Pereira G A S. Parallel sensor-space lattice planner for real-time obstacle avoidance[J]. Sensors, 2022, 22(13): 4770. doi: 10.3390/s22134770 URL
[11]	XU Shaobing, Zidek R, CAO Zhong, et al. System and experiments of model-driven motion planning and control for autonomous vehicles[J]. IEEE Tran Syst, Man, Cybernet: Syst, 2022, 52(9): 5975-5988.
[12]	Manzinger S, Pek C, Althoff M. Using reachable sets for trajectory planning of automated vehicles[J]. IEEE Trans Intel Vehi, 2021, 6(2): 232-248.
[13]	Rasekhipour Y, Khajepour A, Chen S K, et al. A potential field-based model predictive path-planning controller for autonomous road vehicles[J]. IEEE Trans Intel Transp Syst, 2016, 18: 1255-1267. doi: 10.1109/TITS.2016.2604240 URL
[14]	CHEN Jianyu, Zhan Wei, Tomizuka M. Constrained iterative lqr for on-road autonomous driving motion planning[C]// 2017 IEEE 20th Int’l Conf Intel Transport Syst (ITSC). Yokohama, Japan. 2017.
[15]	Artuñedo A, Villagra J, Godoy J. Real-Time motion planning approach for automated driving in urban environments[J]. IEEE Access, 2019, 7: 180039-180053. doi: 10.1109/Access.6287639 URL
[16]	Martinez J J, Canudas-de-Wit C. A safe longitudinal control for adaptive cruise control and stop-and-go scenarios[J]. IEEE Trans Contr Syst Tech, 2007, 15: 246-258. doi: 10.1109/TCST.2006.886432 URL
[17]	Svensson L, Eriksson J C. Tuning for ride quality in autonomous vehicle application to linear quadratic path planning algorithm[D]. Uppsala Universitet, Uppsala, Sweden, 2015.
[18]	Sentouh C, Nguyen A, Benloucif M A, et al. Driver-Automation cooperation oriented approach for shared control of lane keeping assist systems[J]. IEEE Trans Contr Syst Tech, 2019, 27: 1962-1978. doi: 10.1109/TCST.87 URL
[19]	Medina-Lee J, Artuñedo A, Godoy J, et al. Merit-Based motion planning for autonomous vehicles in urban scenarios[J]. Sensors, 2021, 21(11): 3755. doi: 10.3390/s21113755 URL
[20]	Werling M, Ziegler J, Kammel S, et al. Optimal trajectory generation for dynamic street scenarios in a Frenet frame[C] // 2010 IEEE Int’l Conf Robot Auto. Anchorage, AK, USA, 2010.

决策变量	轨迹性能指标	公式
纵向舒适性	纵向加速度平均值	$\overline{a_x(s)}$
纵向舒适性	纵向加加速度平均值	$\overline{j_x(s)}$
横向舒适性	横向加速度平均值	$\overline{a_y(s)}$
横向舒适性	横向加加速度平均值	$\overline{j_y(s)}$
安全性	车道偏离距离	$\int_{s_0}^{s_f}\mathrm{d}_i(s)\mathrm{d}s$
效用性	正向纵向加速度	$\overline{\max{(a_x(s),0)}}$
效用性	平均速度	$\overline{v_s}$

决策变量	轨迹性能指标	公式
纵向舒适性	纵向加速度平均值	$\overline{a_x(s)}$
纵向舒适性	纵向加加速度平均值	$\overline{j_x(s)}$
横向舒适性	横向加速度平均值	$\overline{a_y(s)}$
横向舒适性	横向加加速度平均值	$\overline{j_y(s)}$
安全性	车道偏离距离	$\int_{s_0}^{s_f}\mathrm{d}_i(s)\mathrm{d}s$
效用性	正向纵向加速度	$\overline{\max{(a_x(s),0)}}$
效用性	平均速度	$\overline{v_s}$

决策变量	实验1	实验2	实验3	实验4	实验5
纵向舒适性	1.0	1.0	1.0	1.0	0.5
横向舒适性	0.1	1.0	1.0	1.0	0.5
安全性	0.1	0.1	1.0	1.0	0.5
效用性	0.1	0.1	0.1	1.0	0.5

决策变量	实验1	实验2	实验3	实验4	实验5
纵向舒适性	1.0	1.0	1.0	1.0	0.5
横向舒适性	0.1	1.0	1.0	1.0	0.5
安全性	0.1	0.1	1.0	1.0	0.5
效用性	0.1	0.1	0.1	1.0	0.5

最大横向加速度，a_max,_y	3.0 m/s²
最大纵向加速度，a_max,_x	3.0 m/s²
最小纵向加速度，a_min,_x	-5.0 m/s²
最大限制速度，v_limit	15.0 m/s
最小跟车距离，d_c	10.0 m
安全名义距离，d_o	45.0 m
最大制动加速度，a_max,dec	-5.0 m/s²
最大速度，v_max	15.0 m/s