华南理工大学学报(自然科学版) ›› 2023, Vol. 51 ›› Issue (10): 99-109.doi: 10.12141/j.issn.1000-565X.230148

所属专题: 2023绿色智慧交通系统专辑

• 绿色智慧交通系统 • 上一篇    下一篇

基于几何路网结构和强化学习的车辆重定位策略

许伦辉1 余佳芯1 裴明阳1 吴攀2 李鹏3   

  1. 1.华南理工大学 土木与交通学院,广东 广州 510640
    2.重庆交通大学 交通运输学院,重庆 400000
    3.深圳职业技术学院 汽车与交通学院,广东 深圳 518055
  • 收稿日期:2023-03-28 出版日期:2023-10-25 发布日期:2023-06-05
  • 通信作者: 裴明阳(1993-),女,博士,副教授,主要从事交通系统优化建模研究。 E-mail:mingyang@scut.edu.cn
  • 作者简介:许伦辉(1965-),男,博士,教授,主要从事智能交通系统研究。E-mail:lhxu@scut.edu.cn
  • 基金资助:
    广东省基础与应用基础研究基金(2020A1515111024)

Repositioning Strategy for Ride-Hailing Vehicles Based on Geometric Road Network Structure and Reinforcement Learning

XU Lunhui1 YU Jiaxin1 PEI Mingyang1 WU Pan2 LI Peng3   

  1. 1.School of Civil Engineering and Transportation,South China University of Technology,Guangzhou 510640,Guangdong,China
    2.School of Transportation Engineering,Chongqing Jiaotong University,Chongqing 400000,China
    3.School of Automobile and Transportation,Shenzhen Polytechnic,Shenzhen 518055,Guangdong,China
  • Received:2023-03-28 Online:2023-10-25 Published:2023-06-05
  • Contact: 裴明阳(1993-),女,博士,副教授,主要从事交通系统优化建模研究。 E-mail:mingyang@scut.edu.cn
  • About author:许伦辉(1965-),男,博士,教授,主要从事智能交通系统研究。E-mail:lhxu@scut.edu.cn
  • Supported by:
    the Basic and Applied Research Foundation of Guangdong Province(2020A1515111024)

摘要:

网约车司机和乘客双向搜索效率低、准确性差,造成了需求与供应之间的不匹配。网约车重定位策略将车辆提前调度到未来有需求的地区,提高了供需匹配度。现有的研究大多以网络栅格表示城市道路环境,缺少几何拓扑信息,影响了调度准确性。针对这一现象,提出一种基于图神经网络(GNN)和执行者-评论者强化学习算法(A2C)的网约车重定位算法GA2C。该算法学习过程更平稳且能够高维采样,适用于海量网约车进行多智能体最佳重定位策略的学习,并且使用几何路网结构表示城市道路环境,可以将GNN作为函数逼近器学习路网几何信息,此外,引入基于动作价值函数的动作采样策略,增加了动作选择的随机性,从而有效防止竞争。基于Python构建的网约车重定位仿真实验结果如下:GA2C算法的订单响应率为84.2%,显著高于所有对比实验结果;在订单分布对比实验结果中,GA2C在均匀分布、中心状布局、块状布局和棋盘状布局所对应的相对提升分别为1.17%、6.02%、13.12%和14.55%。上述实验结果表明,GA2C算法能够有效对网约车进行重定位,当订单分布呈现明显差异性,且不同需求区域之间距离较近时,能够更好的学习动态需求变化,通过重定位网约车获得最大订单响应率。

关键词: 城市交通, 网约车重定位策略, 多智能体强化学习, 图神经网络, 马尔可夫决策过程

Abstract:

The inefficient and inaccurate bidirectional search by both ride-hailing drivers and passengers leads to a mismatch between supply and demand. Ride-hailing vehicle repositioning strategy can pre-dispatch vehicles to areas with future demand, improving supply-demand matching. However, existing research mostly uses network grids to represent the urban road environment, lacking geometric topological information and reducing the dispatch accuracy. To address this issue, a ride-hailing vehicle relocation algorithm called GA2C was proposed based on Graph Neural Networks (GNN) and Actor-Critic reinforcement learning algorithm. This algorithm has a smoother learning process and can perform high-dimensional sampling, and it is suitable for learning the best relocation strategy for a large number of ride-hailing vehicles as multi-agent systems. Moreover, the geometric network structure was used to represent the urban road environment by using a GNN as a function approximator to learn the geometric information of the road network. Additionally, an action sampling strategy based on action value function was introduced to increase the randomness of action selection, effectively preventing competition. A ride-hailing vehicle relocation simulation experiment was conducted using Python, and the results are as follows: (i) the order response rate of the GA2C algorithm is 84.2%, significantly higher than all the comparative experimental results; (ii) in the order distribution comparative experiment, GA2C’s relative improvements in uniform distribution, central distribution layout, block distribution layout, and checkerboard distribution layout are 1.17%, 6.02%, 13.12%, and 14.55%, respectively. The above experimental results demonstrate that the GA2C algorithm can effectively relocate ride-hailing vehicles. When the order distribution presents significant differences, and the distance between different demand areas is relatively close, it can better learn dynamic demand changes, and achieve maximum order response rate by relocating ride-hailing vehicles.

Key words: urban traffic, repositioning strategy of ride-hailing, multi-agent reinforcement learning, graph neural network, Markov decision process

中图分类号: