0 引言
1 问题描述
1.1 多智能体交通系统解构
1.2 个性化出行界定
2 路径选择模型及博弈协商机制
2.1 车辆行驶模型
2.2 车辆路径决策模型
表1 路径信息权重及其变化范围 |
| 效用评价指标 | 路网管理者 智能体 | 网联车辆驾驶员 智能体 |
|---|---|---|
| 行程时间 | 0 | 0.40±0.10 |
| 起讫点区间距离 | 0 | 0.15±0.05 |
| 起讫点区间交叉口数量 | 0 | 0.10±0.05 |
| 路径饱和度 | 0.55±0.10 | 0.15±0.10 |
| 路段速度比率 | 0.45±0.05 | 0.20±0.10 |
考虑驾驶员个性的博弈协商机制与出行路径选择研究
|
刘捷(1984—),男,内蒙古乌海人,工程师,研究方向为智慧交通、新型材料等。E-mail: 115910079@qq.com |
收稿日期: 2022-07-18
网络出版日期: 2023-03-08
Game Negotiation Mechanism and Travel Route Choice Considering Driver′s Personality
Received date: 2022-07-18
Online published: 2023-03-08
为解决网联环境下驾驶员个性化和系统最优之间的冲突而导致的交通拥堵问题,采用博弈论、效用理论和多智能体建模技术,对驾驶员、信息发布单元以及路网管理者建立多智能体仿真模型,并提出了一种路网管理者与驾驶员之间的博弈协商机制。为了验证该方法的有效性,基于Net-Logo多智能体仿真软件,搭建了网联交通系统仿真平台,对不同路网饱和度和不同类型驾驶员比例下的出行路径信息服务策略进行仿真。仿真结果表明:所有实验的博弈协商成功率均达到90%以上,路网管理者的系统最优策略与驾驶员的个性化策略得到充分结合;当路网饱和度为1、个性化驾驶员比例为20%时,博弈协商成功率达到98%,博弈协商机制效果最明显。
刘捷 , 王曈 , 孙恒飞 , 沐波 . 考虑驾驶员个性的博弈协商机制与出行路径选择研究[J]. 交通运输研究, 2023 , 9(1) : 86 -95 . DOI: 10.16503/j.cnki.2095-9931.2023.01.009
To solve the problem of traffic congestion caused by the conflict between driver personalization and system optimization in a connected environment, a multi-agent simulation model was established for drivers, information release units, and road network managers by using game theory, utility theory, and multi-agent modeling technology. Then a game negotiation mechanism between road network managers and drivers was proposed. To verify the effectiveness of the proposed method, a connected traffic system simulation platform was built based on Net-Logo multi-agent simulation software, and the travel routing information service strategies under different road network saturation and different types of driver proportions were simulated and verified. The simulation results show that the success rate of game negotiation in all experiments is more than 90%, the network manager′s system optimal strategy is fully combined with the driver′s personalized strategy. The success rate of game negotiation reaches 98% when the road network saturation is 1 and the proportion of personalized drivers is 20%, and the effect of the game negotiation mechanism is the most obvious.
表1 路径信息权重及其变化范围 |
| 效用评价指标 | 路网管理者 智能体 | 网联车辆驾驶员 智能体 |
|---|---|---|
| 行程时间 | 0 | 0.40±0.10 |
| 起讫点区间距离 | 0 | 0.15±0.05 |
| 起讫点区间交叉口数量 | 0 | 0.10±0.05 |
| 路径饱和度 | 0.55±0.10 | 0.15±0.10 |
| 路段速度比率 | 0.45±0.05 | 0.20±0.10 |
| [1] |
吕丹丹. 驾驶员路径选择行为建模及计算[D]. 青岛: 山东理工大学, 2012.
|
| [2] |
|
| [3] |
|
| [4] |
|
| [5] |
|
| [6] |
|
| [7] |
|
| [8] |
|
| [9] |
|
| [10] |
|
| [11] |
|
| [12] |
|
| [13] |
|
| [14] |
|
| [15] |
|
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
陈秀锋, 陈伟. 诱导信息条件下驾驶路径选择行为分析[J]. 现代电子技术, 2022, 45(1): 132-135.
|
| [20] |
李建敏. 驾驶员路径选择行为的演化博弈研究[D]. 大连: 大连海事大学, 2019.
|
| [21] |
孙笑宇. 大数据环境下车联网个性化信息服务模式研究[D]. 长春: 吉林大学, 2016.
|
| [22] |
马力, 施树明. 基于车联网的驾驶博弈行为仿真[C]// 2016中国汽车工程学会年会论文集. 上海: 机械工业出版社, 2016:1989-1994.
|
| [23] |
安实, 崔娜, 于航. 基于Multi-agent协商的出行信息个性化服务策略[J]. 西南交通大学学报, 2010, 45(4):627-634.
|
| [24] |
崔娜. 基于多智能体协商的驾驶员路径选择行为仿真研究[D]. 哈尔滨: 哈尔滨工业大学, 2007.
|
| [25] |
|
| [26] |
赵春晓, 魏楚元. 多智能体系统建模、仿真及应用[M]. 北京: 中国水利水电出版社, 2021:1-9.
|
| [27] |
任福田, 刘小明, 荣建, 等. 交通工程学[M]. 北京: 人民交通出版社, 2008: 113-119.
|
| [28] |
|
/
| 〈 |
|
〉 |