1、 本科生毕业设计 （论文） 外 文文献翻译 毕业设计题目： 交通灯智能控制系统 学 院： 信息科学与工程学院 专业班级： 测控技术与仪器 0703 班 学生姓名： 王欣 指导教师： 桑海峰 2011 年 3 月 19 日 外文 原文 Intelligent Traffic Light Control Marco Wiering, Jelle van Veenen, Jilles Vreeken, and Arne Koopman Intelligent Systems Group Institute of Information and Computing Sciences Utrecht U
2、niversity Padualaan 14, 3508TB Utrecht, The Netherlands email: marcocs.uu.nl July 9, 2004 Abstract Vehicular travel is increasing throughout the world, particularly in large urban areas.Therefore the need arises for simulating and optimizing traffic control algorithms to better accommodate this incr
3、easing demand. In this paper we study the simulation and optimization of traffic light controllers in a city and present an adaptive optimization algorithm based on reinforcement learning. We have implemented a traffic light simulator, Green Light District, that allows us to experiment with differen
4、t infrastructures and to compare different traffic light controllers. Experimental results indicate that our adaptive traffic light controllers outperform other fixed controllers on all studied infrastructures. Keywords: Intelligent Traffic Light Control, Reinforcement Learning, Multi-Agent Systems
5、(MAS), Smart Infrastructures, Transportation Research 1 Introduction Transportation research has the goal to optimize transportation flow of people and goods.As the number of road users constantly increases, and resources provided by current infrastructures are limited, intelligent control of traffi
6、c will become a very important issue in the future. However, some limitations to the usage of intelligent traffic control exist. Avoiding traffic jams for example is thought to be beneficial to both environment and economy, but improved traffic-flow may also lead to an increase in demand Levinson, 2
7、003. There are several models for traffic simulation. In our research we focus on microscopic models that model the behavior of individual vehicles, and thereby can simulate dynamics of groups of vehicles. Research has shown that such models yield realistic behavior Nagel and Schreckenberg, 1992, Wa
8、hle and Schreckenberg, 2001. Cars in urban traffic can experience long travel times due to inefficient traffic light control. Optimal control of traffic lights using sophisticated sensors and intelligent optimization algorithms might therefore be very beneficial. Optimization of traffic light switch
9、ing increases road capacity and traffic flow, and can prevent traffic congestions. Traffic light control is a complex optimization problem and several intelligent algorithms, such as fuzzy logic, evolutionary algorithms, and reinforcement learning (RL) have already been used in attempts to solve it.
10、 In this paper we describe a model-based, multi-agent reinforcement learning algorithm for controlling traffic lights. In our approach, reinforcement learning Sutton and Barto, 1998, Kaelbling et al., 1996 with road-user-based value functions Wiering, 2000 is used to determine optimal decisions for
11、each traffic light. The decision is based on a cumulative vote of all road users standing for a traffic junction, where each car votes using its estimated advantage (or gain) of setting its light to green. The gain-value is the difference between the total time it expects to wait during the rest of
12、its trip if the light for which it is currently standing is red, and if it is green. The waiting time until cars arrive at their destination is estimated by monitoring cars flowing through the infrastructure and using reinforcement learning (RL) algorithms. We compare the performance of our model-ba
13、sed RL method to that of other controllers using the Green Light District simulator (GLD). GLD is a traffic simulator that allows us to design arbitrary infrastructures and traffic patterns, monitor traffic flow statistics such as average waiting times, and test different traffic light controllers.
14、The experimental results show that in crowded traffic, the RL controllers outperform all other tested non-adaptive controllers. We also test the use of the learned average waiting times for choosing routes of cars through the city (co-learning), and show that by using co-learning road users can avoi
15、d bottlenecks. This paper is organized as follows. Section 2 describes how traffic can be modelled, predicted, and controlled. In section 3 reinforcement learning is explained and some of its applications are shown. Section 4 surveys several previous approaches to traffic light control, and introduces our new algorithm. Section 5 describes the
