基于滚动时域优化的动态车队协同驻站控制策略

赖心荷; 徐婷; 姜瑞森; 邓恺龙; 张梁颢; 张志顺

doi:10.19818/j.cnki.1671-1637.2025.06.019

基于滚动时域优化的动态车队协同驻站控制策略

doi: 10.19818/j.cnki.1671-1637.2025.06.019

长安大学运输工程学院，陕西西安 710064

基金项目:

国家重点研发计划 2018YFC0807500

陕西省自然科学基础研究计划项目 2021JZ-20

陕西省自然科学基础研究计划项目 2025JC-YBMS-538

详细信息

作者简介:
赖心荷(2000-)，女，陕西安康人，长安大学工学博士研究生，从事交通运输规划、城市系统目标优化研究

通讯作者:
徐婷(1983-)，女，江苏常州人，长安大学教授，工学博士

中图分类号: U491.1
计量
- 文章访问数: 114
- HTML全文浏览量: 61
- PDF下载量: 13
- 被引次数: 0
出版历程
- 收稿日期: 2024-04-29
- 录用日期: 2025-04-30
- 修回日期: 2025-02-25
- 刊出日期: 2025-12-28

Dynamic fleet cooperative holding control strategy based on rolling horizon optimization

School of Transportation Engineering, Chang'an University, Xi'an 710064, Shaanxi, China

Funds:

National Key R&D Program of China 2018YFC0807500

Natural Science Basic Research Program of Shaanxi Province 2021JZ-20

Natural Science Basic Research Program of Shaanxi Province 2025JC-YBMS-538

More Information

Corresponding author: XU Ting (1983-), female, professor, PhD, annabelxu@163.com

Article Text (Baidu Translation)

摘要

摘要: 引入车队协同优化机制，构建了以车头时距偏差和乘客车内等待时间为目标的车队协同驻站控制策略优化模型；设计了基于自适应遗传算法(AGA)和多起点邻域搜索算法(MNS)的混合启发式算法(AGA-MNS)对控制策略优化模型进行求解；通过西安市实际数据对控制结果进行分析，讨论了最大驻站控制时间以及车头时距偏差权重等因素对驻站控制策略的影响。研究结果表明：基于滚动时域优化的动态车队协同驻站控制策略能够减少37.4%的车头时距偏差和乘客车内等待时间，相比于传统单车驻站控制策略，基于滚动时域优化的动态车队协同驻站控制策略的优化比例提高了8.19%；相比于AGA，AGA-MNS的求解时间减少了13.03%，求解结果标准差降低了0.07 min；随着最大驻站时间的增加，目标函数值持续减少，当最大驻站控制时间大于29 s后目标函数保持稳定；随着车头时距偏差权重的增加，控制策略会设置更长的驻站控制时间以保证车头时距稳定性，当车内等待时间权重等于0.5时，控制策略会设置驻站控制时间使系统目标函数最小。
- 智能交通系统 /
- 动态车队 /
- 公交驻站控制策略 /
- 滚动时域优化 /
- 车头时距稳定性 /
- 混合启发式求解算法
Abstract: A fleet coordination optimization mechanism was introduced, upon which an optimization model for cooperative holding control was formulated with headway deviation and in-vehicle passenger waiting time as the objectives. A hybrid heuristic algorithm (AGA-MNS) based on adaptive genetic algorithm (AGA) and multi-start neighborhood search (MNS) was proposed to solve the optimization model. The control results were analyzed by real-world data in Xi'an, and the effects of the holding strategy with the maximum holding times and the weighting of the headway deviation on the holding control strategy were discussed. Research results indicate that the dynamic fleet cooperative holding control strategy based on receding horizon optimization reduced headway deviation and in-vehicle passenger waiting time by 37.4%. Compared to the traditional holding strategy, the optimization ratio of the dynamic fleet cooperative holding strategy based on rolling horizon optimization is increased by 8.19%. Compared to AGA, the solution time for AGA-MNS is reduced by 13.03%, and the standard deviation of the solution result is reduced by 0.07 min. As the maximum holding times increase, the objective function value continues to decrease and remains stable when the maximum resident control time is greater than 29 s. As the weighting of the headway deviation increases, the strategy improves the holding time to ensure the stability of the headway. When the weighting of the headway deviation equals 0.5, the holding strategy sets the resident holding times to minimize the objective function value.
- intelligent transportation system /
- dynamic fleet /
- bus holding control strategy /
- rolling horizon optimization /
- headway stability /
- hybrid heuristic algorithm

HTML全文

图 1 基于滚动时域优化的动态车队协同驻站控制策略

Figure 1. Dynamic fleet cooperative holding control strategy based on rolling horizon optimization

下载: 全尺寸图片幻灯片

图 2 AGA-MNS流程

Figure 2. Flow of AGA-MNS

下载: 全尺寸图片幻灯片

图 3 染色体编码

Figure 3. Chromosome coding

下载: 全尺寸图片幻灯片

图 4 AGA-MNS多起点邻域搜索流程

Figure 4. Flow of multi-start neighborhood search in AGA-MNS

下载: 全尺寸图片幻灯片

图 5 研究线路轨迹

Figure 5. Trajectory of study routes

下载: 全尺寸图片幻灯片

图 6 控制策略优化前后轨迹对比

Figure 6. Trajectory comparison before and after holding control strategy optimization

下载: 全尺寸图片幻灯片

图 7 不同驻站控制策略控制结果

Figure 7. Control results of various holding control strategies

下载: 全尺寸图片幻灯片

图 8 不同启发式算法迭代过程

Figure 8. Iterations process of various heuristic algorithms

下载: 全尺寸图片幻灯片

图 9 控制策略敏感性分析

Figure 9. Sensitivity analysis of holding control strategy

下载: 全尺寸图片幻灯片

表 1 研究线路乘客特征

Table 1. Passenger characteristics of study routes

站点序号	乘客到达率/(人次·min^-1)		下车比例/%		仅能乘坐本线路车辆的乘客比例/%		站点下车人数中换乘比例/%
站点序号	线路1	线路2	线路1	线路2	线路1	线路2	线路1	线路2
1	1.11	1.53	0.0	0.0	100.0	100.0	0.0	0.0
2	1.25	1.30	10.4	7.6	100.0	100.0	0.0	0.0
3	1.25	1.72	11.3	5.7	100.0	100.0	0.0	0.0
4	0.42	1.57	9.4	6.8	100.0	100.0	0.0	0.0
5	0.18	0.55	11.3	8.4	100.0	100.0	0.0	0.0
6	0.71	1.57	21.2	11.7	100.0	100.0	0.0	0.0
7	0.35	1.46	33.5	12.5	100.0	100.0	0.0	0.0
8	0.73	1.53	18.4	25.4	100.0	100.0	0.0	0.0
9	1.25	1.13	32.1	32.3	100.0	100.0	0.0	0.0
10	1.19	1.11	19.5	32.4	100.0	0.0	0.0	17.9
11	1.08	1.03	28.6	37.4	100.0	0.0	0.0	5.3
12	0.87	0.92	34.2	23.5	100.0	0.0	0.0	2.3
13	1.25	0.92	32.7	42.1	63.4	0.0	0.0	2.7
14	0.75	0.72	29.1	27.5	56.7	0.0	0.0	1.3
15	0.78	0.83	26.4	21.2	58.2	0.0	0.0	3.7
16	1.15	0.76	36.8	28.3	42.6	0.0	0.0	2.8
17	0.55	0.60	28.2	25.4	37.3	0.0	0.0	1.5
18	0.69	0.78	19.4	58.4	33.2	0.0	0.0	1.6
19	0.75	0.57	31.7	55.3	25.7	0.0	0.0	1.3
20	1.71	0.20	22.1	48.5	12.4	0.0	0.0	1.1
21	1.85	0.00	32.3	100.0	3.7		0.0
22	0.62		39.5		0.1		0.0
23	1.11		47.6		0.1		0.0
24	1.42		48.4		100.0		0.0
25	0.82		37.4		100.0		0.0
26	0.95		58.2		100.0		0.0
27	0.49		47.9		100.0		0.0
28	0.00		100.0

下载: 导出CSV

表 2 不同驻站控制策略驻站控制时间

Table 2. Holding control time of various holding control strategies min

控制策略	车辆序号
控制策略	1	2	3	4	5	6	7	8	9	10	总计
单车驻站控制策略^[13]	2.17	4.31	3.44	2.88	1.64	5.31	5.20	2.54	1.79	3.11	32.39
车队协同驻站控制策略	2.07	3.99	3.32	2.66	1.59	5.07	4.96	2.47	1.59	2.86	30.57

下载: 导出CSV

表 3 不同算法求解结果比较

Table 3. Comparison of various algorithms results min

滚动时域规模		CPLEX	AGA^[29]			AGA-MNS
车辆	站点	CPLEX	平均值	优化解	标准差	平均值	优化解	标准差
1	1	50.94	50.94	50.94	0.00	50.94	50.94	0.00
1	2	50.81	50.81	50.81	0.00	50.81	50.81	0.00
1	3	50.72	50.72	50.72	0.01	50.72	50.72	0.00
1	4	50.72	50.94	50.72	0.01	50.72	50.72	0.01
1	5	50.63	51.03	50.63	0.02	50.81	50.63	0.02
2	1	48.72	48.72	48.72	0.00	48.72	48.72	0.00
2	2	48.22	48.22	48.22	0.00	48.22	48.22	0.00
2	3	47.44	47.44	47.44	0.00	47.44	47.44	0.00
2	4	47.23	47.61	47.23	0.03	47.31	47.23	0.02
2	5	47.12	47.58	47.12	0.06	47.26	47.12	0.02
3	1	47.63	47.63	47.63	0.00	47.63	47.63	0.00
3	2	47.15	47.15	47.15	0.00	47.15	47.15	0.00
3	3	46.88	47.31	46.88	0.05	47.03	46.88	0.04
3	4	46.72	47.56	46.72	0.08	47.02	46.72	0.05
3	5	46.69	46.93	46.69	0.10	46.89	46.69	0.08
4	1	47.61	47.91	47.61	0.03	47.71	47.61	0.01
4	2	47.13	47.53	47.13	0.05	47.33	47.13	0.02
4	3	46.77	46.95	46.77	0.11	46.91	46.77	0.02
4	4	46.68	46.93	46.68	0.22	46.83	46.68	0.03
4	5	46.65	47.81	46.65	0.35	46.77	46.65	0.03
5	1	47.59	48.55	47.59	0.18	47.63	47.59	0.02
5	2	47.08	47.51	47.08	0.11	47.31	47.08	0.05
5	3	46.76	47.42	46.76	0.32	47.22	46.76	0.07
5	4	46.67	47.13	46.71	0.24	46.71	46.67	0.08
5	5	46.65	48.13	46.67	0.41	46.68	46.65	0.08
平均		47.89	48.26	47.89	0.10	47.99	47.89	0.03

下载: 导出CSV

表 4 不同算法求解时间比较

Table 4. Comparison of solution time by various algorithms

滚动时域规模		CPLEX/s	AGA^[29]/s	AGA-MNS/s
车辆	站点	CPLEX/s	AGA^[29]/s	AGA-MNS/s
1	1	41.35	13.05	11.20
1	2	59.65	14.20	11.80
1	3	86.70	17.02	17.00
1	4	90.75	19.80	17.10
1	5	109.65	20.75	20.95
2	1	56.71	13.27	13.11
2	2	83.38	16.23	15.79
2	3	93.18	18.89	17.22
2	4	110.39	22.21	19.37
2	5	135.68	25.22	23.17
3	1	86.39	14.99	12.57
3	2	96.11	19.38	17.89
3	3	122.53	20.91	18.23
3	4	144.69	23.21	21.52
3	5	164.26	27.27	23.54
4	1	93.22	19.79	17.30
4	2	115.69	22.39	19.11
4	3	148.22	23.18	21.22
4	4	169.11	27.19	23.77
4	5	195.13	34.56	28.11
5	1	112.61	20.11	20.42
5	2	135.77	24.53	23.58
5	3	159.31	27.16	23.27
5	4	198.77	35.11	29.33
5	5	211.35	41.12	30.22
平均		120.82	22.46	19.87

下载: 导出CSV

参考文献(32)

[1]	KHAN Z S, HE W L, MENENDEZ M. Application of modular vehicle technology to mitigate bus bunching[J]. Transportation Research Part C: Emerging Technologies, 2023, 146: 103953. doi: 10.1016/j.trc.2022.103953
[2]	徐猛, 刘涛, 钟绍鹏, 等. 城市智慧公交研究综述与展望[J]. 交通运输系统工程与信息, 2022, 22(2): 91-108. XU Meng, LIU Tao, ZHONG Shao-peng, et al. Urban smart public transport studies: A review and prospect[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(2): 91-108.
[3]	BERREBI S J, HANS E, CHIABAUT N, et al. Comparing bus holding methods with and without real-time predictions[J]. Transportation Research Part C: Emerging Technologies, 2018, 87: 197-211. doi: 10.1016/j.trc.2017.07.012
[4]	姚顽强. 基于GPS及GIS技术的城市公共交通调度系统[J]. 长安大学学报(自然科学版), 2008, 28(5): 99-102. YAO Wan-qiang. Urban public transit dispatching system based on GPS and GIS technologies[J]. Journal of Chang'an University (Natural Science Edition), 2008, 28(5): 99-102.
[5]	黄青霞, 贾斌, 强生杰, 等. 基于驻站和限流的组合公交控制策略研究[J]. 交通运输系统工程与信息, 2018, 18(4): 103-109. HUANG Qing-xia, JIA Bin, QIANG Sheng-jie, et al. Integrated bus control strategy considering holding and limited-boarding[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18(4): 103-109.
[6]	BIE Y M, XIONG X Y, YAN Y D, et al. Dynamic headway control for high-frequency bus line based on speed guidance and intersection signal adjustment[J]. Computer-aided Civil and Infrastructure Engineering, 2020, 35(1): 4-25. doi: 10.1111/mice.12446
[7]	ZIMMERMANN L, KRAUS J W, KOEHLER L A, et al. Holding control of bus bunching without explicit service headways[J]. IFAC-PapersOnLine, 2016, 49(3): 209-214. doi: 10.1016/j.ifacol.2016.07.035
[8]	GKIOTSALITIS K, VAN BERKUM E C. An exact method for the bus dispatching problem in rolling horizons[J]. Transportation Research Part C: Emerging Technologies, 2020, 110: 143-165. doi: 10.1016/j.trc.2019.11.009
[9]	ASGHARZADEH M, SHAFAHI Y. Real-time bus-holding control strategy to reduce passenger waiting time[J]. Transportation Research Record, 2017(2647): 9-16.
[10]	GKIOTSALITIS K. Bus rescheduling in rolling horizons for regularity-based services[J]. Journal of Intelligent Transportation Systems, 2021, 25(4): 356-375. doi: 10.1080/15472450.2019.1681992
[11]	DE SOUZA F, TEIXEIRA S M. Improving resilience of bus bunching holding strategy through a rolling horizon approach[J]. Journal of Transportation Engineering, Part A: Systems, 2021, 147(10): 04021074. doi: 10.1061/JTEPBS.0000587
[12]	MIRZAEI M, SHEIKHOLESLAMI A. Bus holding control strategy at all the rolling horizon stops under possible and impossible overtaking conditions[J]. Journal of Transportation Research, 2024(6): 493-506.
[13]	DRABICKI A, KUCHARSKI R, CATS O. Mitigating bus bunching with real-time crowding information[J]. Transportation, 2023, 50(3): 1003-1030. doi: 10.1007/s11116-022-10270-3
[14]	BERREBI S J, WATKINS K E, LAVAL J A. A real-time bus dispatching policy to minimize passenger wait on a high frequency route[J]. Transportation Research Part B: Methodological, 2015, 81: 377-389. doi: 10.1016/j.trb.2015.05.012
[15]	SAEED Z, WEILI H, MÓNICA M. Application of modular vehicle technology to mitigate bus bunching[J]. Transportation Research Part C: Emerging Technologies, 2023, 146: 103953. doi: 10.1016/j.trc.2022.103953
[16]	赵琥, 冯树民, 廖嘉雯, 等. 车路协同环境下重叠线路公交车速诱导策略[J]. 中国公路学报, 2021, 34(7): 42-53. ZHAO Hu, FENG Shu-min, LIAO Jia-wen, et al. Speed guidance strategy for buses run on overlapping route segments under cooperative vehicle infrastructure[J]. China Journal of Highway and Transport, 2021, 34(7): 42-53.
[17]	XUAN Y G, ARGOTE J, DAGANZO C F. Dynamic bus holding strategies for schedule reliability: Optimal linear control and performance analysis[J]. Transportation Research Part B: Methodological, 2011, 45(10): 1831-1845. doi: 10.1016/j.trb.2011.07.009
[18]	董长印, 熊卓智, 李霓, 等. 考虑减缓交通振荡的混合队列控制方法[J]. 交通运输工程学报, 2024, 24(6): 212-229. DONG Chang-yin, XIONG Zhuo-zhi, LI Ni, et al. Mixed platoon control method considering traffic oscillation mitigation[J]. Journal of Traffic and Transportation Engineering, 2024, 24(6): 212-229.
[19]	谢宪毅, 赵鑫, 金立生, 等. 融合深度强化学习与滚动时域优化的智能车辆轨迹跟踪控制[J]. 交通运输工程学报, 2024, 24(6): 259-272. doi: 10.19818/j.cnki.1671-1637.2024.06.018 XIE Xian-yi, ZHAO Xin, JIN Li-sheng, et al. Trajectory tracking control of intelligent vehicles based on deep reinforcement learning and rolling horizon optimization[J]. Journal of Traffic and Transportation Engineering, 2024, 24(6): 259-272. doi: 10.19818/j.cnki.1671-1637.2024.06.018
[20]	HUANG H F, HUANG L, SONG R J, et al. Bus single-trip time prediction based on ensemble learning[J]. Computational Intelligence and Neuroscience, 2022, DOI: 10.1155/2022/6831167.
[21]	HASSANNAYEBI E, FARJAD A, AZADNIA A, et al. A data analytics framework for reliable bus arrival time prediction using artificial neural networks[J]. International Journal of Data Science and Analytics, 2025, 20: 337-356. doi: 10.1007/s41060-023-00391-y
[22]	ZHANG X Y, RICE J A. Short-term travel time prediction[J]. Transportation Research Part C: Emerging Technologies, 2003, 11(3/4): 187-210.
[23]	AGAFONOV A A, YUMAGANOV A S. Bus arrival time prediction using recurrent neural network with LSTM architecture[J]. Optical Memory and Neural Networks, 2019, 28: 222-230. doi: 10.3103/S1060992X19030081
[24]	YUAN Y, SHAO C F, CAO Z C, et al. Bus dynamic travel time prediction: Using a deep feature extraction framework based on RNN and DNN[J]. Electronics, 2020, 9(11): 1876. doi: 10.3390/electronics9111876
[25]	CUI Z Y, KE R M, PU Z Y, et al. Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values[J]. Transportation Research Part C: Emerging Technologies, 2020, 118: 1-14.
[26]	张兵, 周丹丹, 孙健, 等. 基于双向长短期记忆网络的公交到站时间预测模型[J]. 交通运输系统工程与信息, 2023, 23(2): 148-160. ZHANG Bing, ZHOU Dan-dan, SUN Jian, et al. Bus arrival time prediction model based on bidirectional long short-term memory network[J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(2): 148-160.
[27]	姜瑞森, 胡大伟, 孙倩, 等. 基于实时车辆信息共享的公交动态控制策略[J]. 长安大学学报(自然科学版), 2023, 43(6): 95-105. JIANG Rui-sen, HU Da-wei, SUN Qian, et al. Dynamic control strategy for public transportation based on real-time information sharing[J]. Journal of Chang'an University(Natural Science Edition), 2023, 43(6): 95-105.
[28]	龙雪琴, 李景涛, 王建军, 等. 实时信息下共线公交线路发车时刻表的协同优化[J]. 华南理工大学学报(自然科学版), 2022, 50(2): 23-32. LONG Xue-qin, LI Jing-tao, WANG Jian-jun, et al. Collaborative optimization of departure timetable for common bus lines under real-time information[J]. Journal of South China University of Technology (Natural Science Edition), 2022, 50(2): 23-32.
[29]	SHIMA Y, KADIR R, ALI F. A novel approach to the optimization of a public bus schedule using K-means and a genetic algorithm[J]. IEEE Access, 2021, 9: 73365-73376. doi: 10.1109/ACCESS.2021.3080508
[30]	李翠玉, 胡雅梦, 康亚伟, 等. 应用自适应遗传算法的电动汽车充放电协同调度[J]. 吉林大学学报(工学版), 2022, 52(11): 2508-2513. LI Cui-yu, HU Ya-meng, KANG Ya-wei, et al. Coordination scheduling of electric vehicle charge and discharge using adaptive genetic algorithm[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(11): 2508-2513.
[31]	靳海涛, 金凤君, 陈卓, 等. 基于换乘链断裂点时空信息的公交换乘行为识别[J]. 交通运输工程学报, 2018, 18(5): 176-184. doi: 10.19818/j.cnki.1671-1637.2018.05.017 JIN Hai-tao, JIN Feng-jun, CHEN Zhuo, et al. Commute activity identification based on spatial and temporal information of transit chaining breaks[J]. Journal of Traffic and Transportation Engineering, 2018, 18(5): 176-184. doi: 10.19818/j.cnki.1671-1637.2018.05.017
[32]	李利华, 曹慧琪, 邓亚军, 等. 基于站点群体聚集性客流的公交串车调度优化[J]. 中国公路学报, 2023, 36(2): 203-215. LI Li-hua, CAO Hui-qi, DENG Ya-jun, et al. Optimization of bus bunching scheduling based on group-gathered passenger flow at bus stops[J]. China Journal of Highway and Transport, 2023, 36(2): 203-215.