北京航空航天大学学报 ›› 2020, Vol. 46 ›› Issue (12): 2284-2292.doi: 10.13700/j.bh.1001-5965.2019.0627

• 论文 • 上一篇    下一篇

半自动驾驶公交车辆编组与调度优化

代壮1, 陈汐1, 马晓磊1,2   

  1. 1. 北京航空航天大学 交通科学与工程学院, 北京 100083;
    2. 北京航空航天大学 大数据科学与脑机智能高精尖创新中心, 北京 100083
  • 收稿日期:2019-12-16 发布日期:2020-12-28
  • 通讯作者: 马晓磊 E-mail:xiaolei@buaa.edu.cn
  • 作者简介:代壮,男,博士研究生。主要研究方向:智能网联交通;陈汐,男,博士研究生。主要研究方向:公交调度优化;马晓磊,男,博士,副教授,博士生导师。主要研究方向:智能交通系统。
  • 基金资助:
    国家自然科学基金(U1811463,61773036);北京市自然科学基金(9172011)

Semi-autonomous driving bus platooning and scheduling optimization

DAI Zhuang1, CHEN Xi1, MA Xiaolei1,2   

  1. 1. School of Transportation Science and Engineering, Beihang University, Beijing 100083, China;
    2. Big Data Brain Computing, Beihang University, Beijing 100083, China
  • Received:2019-12-16 Published:2020-12-28

摘要: 半自动驾驶公交车辆编组是指半自动驾驶公交单元通过车联网技术连接在一起,实现车辆协同驾驶和车辆容量动态设计的车辆组织技术。以半自动驾驶公交车辆编组为出发点,建立编组车辆动态运行模型,分析编组车辆到离站时间、乘客上下车过程、车辆容量限制和车载乘客数量变化等。在此基础上,以车辆运营成本和乘客候车时间成本之和为目标函数,以车辆编组大小和发车时刻为决策变量,建立半自动驾驶公交车辆调度优化模型。提出改进的遗传算法高效求解模型。以杭州55路公交线路为实证案例,仿真结果表明:相比于传统人工驾驶公交的车辆调度,基于半自动驾驶公交的车辆调度能降低29.2%的车辆运营成本和18.2%的乘客候车时间成本,所得结果证实了所建模型优化半自动驾驶公交车辆调度的有效性。

关键词: 智能交通系统, 半自动驾驶, 车辆编组, 公交车辆调度, 遗传算法

Abstract: Semi-autonomous driving bus platooning refers to the vehicle organization technology that connects bus units together through vehicle communication technologies to realize coordinated driving of vehicles and dynamic design of vehicle capacity. Based on semi-autonomous driving us platooning, a dynamic bus operation model is first proposed to model bus arrival and departure time at stops, passenger dwelling process, bus capacity constraint and onboard passenger dynamics. On this basis, a semi-autonomous driving bus scheduling optimization model is proposed to jointly optimize platooning size and bus dispatching time with the objective of the sum of the optimizing operating cost and passenger waiting time cost. An improved genetic algorithm is proposed to solve the model efficiently. The model is validated using a real-world example of bus route 55, Hangzhou, China. Simulation results show that the proposed semi-autonomous driving bus scheduling can reduce bus operating cost by 29.2% and reduce passenger waiting time cost by 18.2%, when compared with conventional human-driven bus scheduling. The result verifies the efficiency of the proposed model in scheduling semi-autonomous driving bus.

Key words: intelligent transportation system, semi-autonomous driving, bus platooning, bus scheduling, genetic algorithm

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发