高速公路单点入口匝道RLRM控制方法

王兴举; 高桂凤; 宫城俊彦

doi:10.19818/j.cnki.1671-1637.2012.03.015

高速公路单点入口匝道RLRM控制方法

doi: 10.19818/j.cnki.1671-1637.2012.03.015

王兴举^{1, 2,},
高桂凤^{1, 2},
宫城俊彦³

1.
石家庄铁道大学交通运输学院, 河北石家庄 050043
2.
河北省交通安全与控制实验室, 河北石家庄 050043
3.
东北大学信息科学研究科, 宫城仙台 9808578

基金项目:

国家自然科学基金项目 51008201

河北省自然科学基金项目 E2012210016

河北省高等学校科学研究重点项目 GD2010235

详细信息

作者简介:
王兴举(1978-), 男, 辽宁海城人, 石家庄铁道大学副教授, 工学博士, 从事智能交通研究

中图分类号: U491.54
计量
- 文章访问数: 1040
- HTML全文浏览量: 175
- PDF下载量: 1062
- 被引次数: 0
出版历程
- 收稿日期: 2011-12-27
- 刊出日期: 2012-06-25

RLRM control method of single entrance ramp for highway

WANG Xing-ju^{1, 2
,},
GAO Gui-feng^{1, 2},
MIYAGI T³

1.
School of Traffic and Transportation, Shijiazhuang Tiedao University, Shijiazhuang 050043, Hebei, China
2.
Traffic Safety and Control Laboratory of Hebei Province, Shijiazhuang 050043, Hebei, China
3.
Graduate School of Information Sciences, Tohoku University, Sendai 9808578, Miyagi, Japan

More Information

Author Bio:
WANG Xing-ju (1978-), male, associate professor, PhD, +86-311-87939527, wangxingju@stdu.edu.cn

Article Text (Baidu Translation)

摘要

摘要: 为缓解交通堵塞, 基于人工智能的强化学习理论, 提出了不完全信息下的强化学习单点入口匝道控制方法(RLRM)。基于6个仿真实例, 分别计算了平均速度、平均密度、流出交通量与旅行时间, 比较了无控制、定时控制与RLRM控制的控制效果。仿真结果表明: 在交通量较小的实例1中, 以旅行时间为评价指标, 定时控制与RLRM控制的交通阻塞缓解率分别为-6.25%、-9.38%, 几乎没有控制效果; 在交通量变大的实例3中, 以旅行时间为评价指标, 定时控制与RLRM控制的交通阻塞缓解率分别为-8.19%、3.51%, 匝道控制有一定效果, RLRM控制略优于定时控制; 在交通量最大的实例6中, 以平均速度、平均密度、流出交通量与旅行时间为评价指标, 定时控制的交通阻塞缓解率分别为8.20%、0.39%、18.97%与23.99%, RLRM控制的交通阻塞缓解率分别为18.18%、3.42%、30.65%与44.41%, RLRM控制明显优于定时控制。可见, 交通量越大, RLRM控制效果越明显。
- 交通控制 /
- 匝道 /
- 交通流仿真 /
- 人工智能 /
- 强化学习 /
- RLRM控制
Abstract: In order to relieve freeway traffic congestion, reinforcement learning ramp metering(RLRM) control method for single entrance ramp of highway under the incomplete information was proposed based on the artificial intelligence theories of reinforcement learning.Average speeds, average densities, traffic outflows and travel times of numerical cases 1-6 were calculated, and the control effect of RLRM was compared with no control and fixed-time control.Simulation result shows that in case 1 with the lowest traffic inflow, the congestion relief rates of fixed-time control and RLRM control depending on travel time are-6.25% and-9.38% respectively, which indicates that the control effect is not significant.When the traffic inflow increases in case 3, the congestion relief rates of fixed-time control and RLRM control depending on travel time are-8.19% and 3.51% respectively, which indicates that the control has some effect, and RLRM control performs better than fixed-time control.In case 6 with the highest traffic inflow, the congestion relief rates of fixed-time control are 8.20%, 0.39%, 18.97% and 23.99% respectively, and those of RLRM control are 18.18%, 3.42%, 30.65% and 44.41% taking average speed, average density, traffic outflow and travel time as evaluating indexes respectively, which shows that RLRM control effect is more significant than fixed-time control.So the greater the traffic inflow is, the better the control effect of RLRM is.
- traffic control /
- ramp /
- traffic flow simulation /
- artificial intelligence /
- reinforcement learning /
- RLRM control

HTML全文

图 1 高速公路匝道

Figure 1. Freeway ramp

下载: 全尺寸图片幻灯片

图 2 匝道控制算法

Figure 2. Ramp metering algorithm

下载: 全尺寸图片幻灯片

图 3 实例1中平均速度与交通阻塞缓解率

Figure 3. Average speeds and congestion relief rates in case 1

下载: 全尺寸图片幻灯片

图 4 实例1中平均密度与交通阻塞缓解率

Figure 4. Average densities and congestion relief rates in case 1

下载: 全尺寸图片幻灯片

表 1 控制参数

Table 1. Control parameters

T_max/min	R/(pcu·h^-1)	r_max/(pcu·h^-1)	r_n/(pcu·h^-1)	N_r
5	200	1 100	100	11

下载: 导出CSV

表 2 交通流量参数

Table 2. Traffic flow parameters pcu·h^-1

实例	1	2	3	4	5	6
主线流入交通量	1 200	1 500	1 800	1 800	2 500	2 500
匝道流入交通量	300	300	600	900	600	900
交通总量	1 500	1 800	2 400	2 700	3 100	3 400

下载: 导出CSV

表 3 实例2控制评价指标

Table 3. Control evaluation indexes in case 2

控制类型	主线平均速度/(km·h^-1)	主线平均密度/(pcu·km^-1)	流出交通量/(pcu·h^-1)	旅行时间/s
无控制	108.85	28.32	2 056	95
定时控制	101.61	33.97	1 763	119
RLRM控制	101.79	34.57	2 068	100

下载: 导出CSV

表 4 实例4控制评价指标

Table 4. Control evaluation indexes in case 4

控制类型	主线平均速度/(km·h^-1)	主线平均密度/(pcu·km^-1)	流出交通量/(pcu·h^-1)	旅行时间/s
无控制	63.57	77.78	2 445	503
定时控制	63.73	94.38	2 639	466
RLRM控制	72.18	71.41	2 701	311

下载: 导出CSV

表 5 实例5控制评价指标

Table 5. Control evaluation indexes in case 5

控制类型	主线平均速度/(km·h^-1)	主线平均密度/(pcu·km^-1)	流出交通量/(pcu·h^-1)	旅行时间/s
无控制	53.42	109.33	2 054	558
定时控制	54.86	101.02	2 311	407
RLRM控制	64.37	98.79	3 176	333

下载: 导出CSV

参考文献(16)

[1]	朱建国, 王朝辉. 高速公路建设精细化管理效果评价体系[J]. 长安大学学报: 自然科学版, 2012, 32(2): 52-57. https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL201202012.htm ZHU Jian-guo, WANG Chao-hui. Effect evaluation system of freeway construction meticulous management[J]. Journal of Chang'an University: Natural Science Edition, 2012, 32(2): 52-57. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL201202012.htm
[2]	干宏程, 陈建阳, 汪晴. 高速公路网路径诱导与匝道控制集成[J]. 系统工程, 2009, 27(6): 78-82. https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200906012.htm GAN Hong-cheng, CHEN Jian-yang, WANG Qing. Integrated routing and ramp metering control for freeway networks[J]. Systems Engineering, 2009, 27(6): 78-82. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GCXT200906012.htm
[3]	徐建闽, 李林, 林培群, 等. 高速公路交通管制策略仿真评价[J]. 公路, 2011(2): 83-87. https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201102018.htm XU Jian-min, LI Lin, LIN Pei-qun, et al. Simulation-based evaluation for expressway traffic control strategies[J]. Highway, 2011(2): 83-87. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201102018.htm
[4]	曹志威, 韩印. 基于模糊理论的高速公路入口匝道控制研究[J]. 交通与运输, 2011(7): 54-58. https://www.cnki.com.cn/Article/CJFDTOTAL-JTYH201101016.htm CAO Zhi-wei, HAN Yin. Study on freeway on-ramp control based on fuzzy theory[J]. Traffic and Transportation, 2011(7): 54-58. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JTYH201101016.htm
[5]	GOMES G, HOROWITZ R. Optimal freeway ramp metering using the asymmetric cell transmission model[J]. Transpor-tation Research Part C: Emerging Technologies, 2006, 14(4): 244-262. doi: 10.1016/j.trc.2006.08.001
[6]	庞明宝, 贺国光, 马宁. 基于遗传算法高速公路混沌系统模糊延迟反馈匝道控制[J]. 土木工程学报, 2009, 42(3): 125-131. doi: 10.3321/j.issn:1000-131X.2009.03.021 PANG Ming-bao, HE Guo-guang, MA Ning. On-ramp chaos control of freeway system with fuzzy delay feedback based on genetic algorithm[J]. China Civil Engineering Journal, 2009, 42(3): 125-131. (in Chinese). doi: 10.3321/j.issn:1000-131X.2009.03.021
[7]	张海军, 杨晓光, 张珏. 高速道路入口匝道控制方法综述[J]. 同济大学学报: 自然科学版, 2005, 33(8): 1051-1055, 1082. doi: 10.3321/j.issn:0253-374X.2005.08.012 ZHANG Hai-jun, YANG Xiao-guang, ZHANG Jue. Review of freeway on-ramp metering methodologies[J]. Journal of Tongji University: Natural Science, 2005, 33(8): 1051-1055, 1082. (in Chinese). doi: 10.3321/j.issn:0253-374X.2005.08.012
[8]	姜紫峰, 荆便顺, 韩锡令. 高速公路入口匝道控制的仿真研究[J]. 中国公路学报, 1997, 10(2): 83-89. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL199702014.htm JIANG Zi-feng, JING Bian-shun, HAN Xi-ling. Simulation study of on-ramp control in freeway[J]. China Journal of Highway and Transport, 1997, 10(2): 83-89. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL199702014.htm
[9]	PAPAGEORGIOU M, HADJ-SALEM H, MIDDELHAM F. ALINEA local ramp metering: summary of field results[J]. Transportation Research Record, 1997(1603): 90-98. doi: 10.3141/1603-12
[10]	郑飞, 杜豫川, 孙立军. 基于ALINEA算法快速路入口匝道单点动态控制[J]. 同济大学学报: 自然科学版, 2009, 37(6): 766-771. https://www.cnki.com.cn/Article/CJFDTOTAL-TJDZ200906011.htm ZHENG Fei, DU Yu-chuan, SUN Li-jun. ALINEA-based on-ramp traffic-responsive local-metering strategy for express-way in Shanghai[J]. Journal of Tongji University: Natural Science, 2009, 37(6): 766-771. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-TJDZ200906011.htm
[11]	贺敬凯, 徐建闽. 基于BP神经网络的入口匝道控制器的设计[J]. 华南理工大学学报: 自然科学版, 2002, 30(7): 24-27. https://www.cnki.com.cn/Article/CJFDTOTAL-HNLG200207006.htm HE Jing-kai, XU Jian-min. Design of on-ramp controller based on BP neural networks[J]. Journal of South China University of Technology: Natural Science Edition, 2002, 30(7): 24-27. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HNLG200207006.htm
[12]	刘平, 任建亭. 基于PID神经网络的快速路入口匝道控制[J]. 科学技术与工程, 2011, 11(32): 7961-7966. doi: 10.3969/j.issn.1671-1815.2011.32.027 LIU Ping, REN Jian-ting. Freeway ramp metering based on PID neural network[J]. Science Technology and Engineering, 2011, 11(32): 7961-7966. (in Chinese). doi: 10.3969/j.issn.1671-1815.2011.32.027
[13]	陈德望, 王飞跃, 陈龙. 基于模糊神经网络的城市高速公路入口匝道控制算法[J]. 交通运输工程学报, 2003, 3(2): 100-105. http://transport.chd.edu.cn/article/id/200302017 CHEN De-wang, WANG Fei-yue, CHEN Long. Freeway ramp control algorithm based on neurofuzzy networks[J]. Journal of Traffic and Transportation Engineering, 2003, 3(2): 100-105. (in Chinese). http://transport.chd.edu.cn/article/id/200302017
[14]	赵明, 侯忠生, 晏静文. 受限迭代学习入口匝道控制收敛性分析[J]. 交通运输系统工程与信息, 2011, 11(2): 166-173. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201102025.htm ZHAO Ming, HOU Zhong-sheng, YAN Jing-wen. Convergence analysis of ILC based ramp metering under constraints[J]. Journal of Transportation Systems Engineering and Informa-tion Technology, 2011, 11(2): 166-173. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201102025.htm
[15]	侯忠生, 晏静文. 带有迭代学习前馈的快速路无模型自适应入口匝道控制[J]. 自动化学报, 2009, 35(5): 588-595. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO200905022.htm HOU Zhong-sheng, YAN Jing-wen. Model free adaptive control based freeway ramp metering with feedforward itera-tive learning controller[J]. ACTA Automatica Sinica, 2009, 35(5): 588-595. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO200905022.htm
[16]	王兴举, 宫城俊彦. 强化学习型匝道控制模型的研究[J]. 石家庄铁道大学学报: 自然科学版, 2010, 23(2): 104-108. https://www.cnki.com.cn/Article/CJFDTOTAL-SJZT201002019.htm WANG Xing-ju, MIYAGI T. Reinforcement learning ramp metering[J]. Journal of Shijiazhuang Tiedao University: Natural Science Edition, 2010, 23(2): 104-108. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-SJZT201002019.htm