基于Q学习和动态权重的改进的区域交通信号控制方法  

Promoted Traffic Control Strategy Based on Q Learning and Dynamic Weight

作　　者：张辰[1] 喻剑[1] 何良华[1] ZHANG Chen YU Jian HE Liang-hua(Institute of Electronic and Information, Tongji University, Shanghai 400047, Chin)

机构地区：[1]同济大学电子与信息工程学院,上海400047

出　　处：《计算机科学》2016年第8期171-176,共6页Computer Science

摘　　要：Q学习在交通信号控制中具有广泛的应用。在区域交通中,基于Q学习的传统区域交通信号控制方法通过agent之间互相交流的方式获取周边路口信息,并作出最有利的决策。传统交通控制方法在大部分情况下具有良好的表现。然而,由于其对周边路口拥堵程度的回馈计算不准确,因此在周边路口堵塞程度相差较大时将出现决策失误,从而导致局部热点拥堵。针对该问题进行分析,并以传统的区域交通信号控制方法为基础,提出一种新的基于Q学习和动态权重的改进的区域交通信号控制方法,引入"路口权重"的概念,通过多目标组合法将其应用于回馈计算,且权重随路口实际交通情况动态改变,解决了易陷入局部热点拥堵的问题。应用仿真软件在3种不同的交通状况下进行模拟,结果表明,所提算法在"拥堵"的状况下较传统控制方法具有更突出的表现。Q-Learning is widely used in traffic signal control. In traditional multi-agent traffic signal control policy, agents gain intersection information via network, and make the best control decision. It works well in most cases. But traditional policy has a weakness that the global reward is calculated by simple average. This may cause local block in some cases. This paper introduced a promoted area traffic signal control based on Q learning. ＂Intersection Weight＂ is used in the new calculation method, which varies dynamically according to the real traffic condition. Both traditional and promoted methods were used to experiment. The results show the advantage of the promoted one.

关 键 词：Q学习 区域控制 路口权重 

分 类 号：U491.51[交通运输工程—交通运输规划与管理]