云环境下智能交通控制系统的研究

VIP免费

3.0 侯斌 2024-11-19 10 4 10.37MB 41 页 15积分

侵权投诉

摘要

在当今社会的交通领域，交通拥堵、交通安全以及交通污染问题已成为困扰人

们出行过程中的三大难题，随着城市化进程速度的加快，汽车数量的增加速度已

与道路里程、其他道路交通设施的增加速度严重失衡，从而导致了交通问题越来

越严重。如何对海量交通信息进行高效处理、预测、调度、发布，实现人、车、路之

间的信息共享、协同合作，减少交通拥挤和交通事故，提高交通效率，将是关系

到未来交通信息服务的关键性问题。而云计算技术具有海量信息存储、计算资源统

一管理、并行任务调度等优势，它能够有效地缓解道路交通信息系统所面临的诸

多问题，实时的向出行者们发布动态交通服务信息、提供实时路况动态、预估出行

时间、诱导行车路线等。因此，本文首先介绍了云计算技术、任务调度技术以及智

能交通相关技术，对基于云环境下的智能交通控制系统进行整体分析，着力研究

智能交通控制系统中的实时路况服务信息系统，并对该系统进行整体设计，其中

针对交通信息调度效率不高、延迟时间过长等问题，提出一种基于任务延迟时间

最短与兼顾任务分配公平的改进蚁群算法；针对交通流预测过程中所存在的非线

性、随机性以及模糊性等问题，建立交通流预测云模型。

本文具体研究内容如下：

首先，对云环境下的智能交通系统的核心子系统 -- 实时路况服务系统

（LTSS）进行需求分析，并对其三个功能模块：预处理模块、任务调度模块以及

交通流预测模块进行总体设计。

其次，在任务的调度过程中往往存在效率不高、调度延迟时间过长等问题，因

此在任务调度模块采用基于任务延迟时间最短和任务分配公平的改进蚁群算法

（DSFACO）的云计算任务调度方法。建立了一种既能改善任务并行性，又能兼

顾任务串行关系的并行调度模型，将预处理模块中的任务按运行次序放入具有不

同优先级的调度队列中，针对同一调度队列中的子任务，采用基于任务延迟时间

最短的改进蚁群算法进行调度，该算法在兼顾公平与效率的同时，使得任务延迟

时间大大缩短。通过仿真实验对比分析 DSFACO 算法与 TS-EACO 算法，实验结

果表明 DSFACO 算法在公平性、任务延迟时间和效率方面优于 TS-EACO 算法，

更加适合云环境。

最后，为了给出行者们提供实时准确的交通流信息服务，如提供实时动态路

况、预估出行时间，而在交通流预测过程中往往存在非线性、时变性和不确定性等

问题，因此在交通流预测模块提出了基于云模型环境的短时交通流预测算法，对

交通流做出整体性的处理，利用云模型对交通流进行拟合，通过拟合后的历史云

与当前云，来共同生成预测云以预测当前的交通流量，通过仿真分析短时交通流

量云模型算法，从仿真结果可知该算法能够克服交通流预测所存在的非线性、随

机性以及模糊性等问题，并且具有良好的误差稳定性，为实时路况服务系统提供

实时交通诱导服务提供了可能，进而为出行者们提前避开拥堵的路段，实时选择

规划出行路线创造了很好的契机。

关键词：智能交通控制云计算实时路况服务系统任务调度短时交通

流预测

ABSTRACT

In the field of transport in today's society, traffic congestion, traffic safety and

traffic pollution problems have become the three problems in the process of travel, with

the speeding up of urbanization speed, the speed has been an increase in the number of

cars and the road mileage, imbalance increase speed of road traffic facilities, and other

leading to a traffic problem is more and more serious. How to handle the massive

amounts of traffic information, forecasting, dispatching, distribution, realize

information sharing between people, vehicles, roads, collaboration, reduce traffic

congestion and traffic accidents, improve the efficiency of traffic, will be related to the

traffic information service of the key issues in the future. And cloud computing

technology, with its mass storage, unified management of resources, the advantages of

the parallel task scheduling, can effectively solve the traffic information system faces

many problems, the real-time dynamic traffic information service to the travelers,

provide real-time traffic dynamic, estimated travel time, inducing the route, etc.

Therefore, this paper based on the cloud environment of intelligent traffic control

system is analyzed, the paper introduces the technology of cloud computing, task

scheduling and intelligent transportation technology, the research on intelligent traffic

control system of real-time traffic information service system, and carries on the system

overall design, including for traffic information scheduling efficiency is not high, the

problem such as delay time is too long, in this paper, a delay time shortest based on task

and task allocation fairness of improved ant colony algorithm; According to traffic flow

forecasting by problems such as nonlinear, randomness and fuzziness, traffic flow

forecasting cloud model is established.

The main works of this paper are:

Firstly, analyze the core subsystem of the intelligent transportation system in cloud

environment -- real-time traffic service system (LTSS), and overall design the three

functional modules: pre-processing module, task scheduling module and traffic flow

forecasting module.

Secondly, in the process of the task scheduling, there are some problems such as

the scheduling is inefficient, the scheduling delay time is too long, so in task scheduling

module, the improved ant algorithm based on delay time shortest and task allocation

fairness (DSFACO) is applied to schedule. A parallel scheduling model is proposed,

which can improve the task parallelism while maintaining the serial relationship

between tasks. Put the tasks of pre-processing module into scheduling queue with

different priorities according to running order. For these tasks in the same priority

scheduling queue, the improved ant algorithm based on delay time shortest (DSACO) is

applied to schedule. Considering both fairness and efficiency, DSACO algorithm

greatly reduces task delay time. Through the simulation experiments to analysis the

model algorithm with the TS-EACO algorithm. The experimental results show the

model is better than the TS-EACO algorithm in fairness, efficiency and task delay

time， and better suitable for the cloud computing scheduling.

Finally, in order to provide accurate real-time traffic flow information services for

travelers: to provide real-time dynamic traffic, estimate the travel time, but in the

process of traffic flow forecasting, there are the problems of nonlinear, time-varying and

uncertainty. So the short-term traffic flow prediction algorithm based on cloud model is

proposed in traffic flow forecasting module, do holistic treat with traffic flow and use

cloud model to fit traffic, use historical and the current traffic flow to establish history

and current cloud and together generate predicted cloud to predict traffic flow. The

simulation shows that short-term traffic flow forecasting method based on cloud model

has high prediction accuracy, it not only can solve the problems of randomness,

fuzziness and nonlinear existence in short-term traffic flow forecasting, but also can be

very good to avoid noise problems caused by the prediction error.

Keywords: Intelligent Traffic Control, Cloud Computing, Live traffic

service system, Task scheduling, Short-term Traffic Flow Prediction

目  录
中文摘要
ABSTRACT
第一章 绪论.............................................................................................................1
§1.1 研究背景与意义..........................................................................................1
§1.2 国内外研究现状..........................................................................................2
§1.2.1 云计算环境研究现状...........................................................................2
§1.2.2 智能交通系统国内外研究现状...........................................................3
§1.3 主要研究工作及结构安排..........................................................................5
第二章 云环境下的智能交通控制系统相关技术研究.........................................7
§2.1 云计算和任务调度技术..............................................................................7
§2.3 智能交通相关技术研究..............................................................................9
§2.4 本章小结....................................................................................................10
第三章 云环境下的智能交通控制系统设计.......................................................11
§3.1 云环境下的智能交通控制系统设计分析................................................11
§3.2 云环境下的智能交通控制系统整体模型设计........................................12
§3.3 云环境下的系统功能模块设计................................................................14
§3.3.1 预处理模块设计.................................................................................14
§3.3.2 任务调度模块设计.............................................................................14
§3.3.3 交通流预测模块设计.........................................................................15
§3.4 云环境下动态交通信息的发布................................................................15
§3.5 云环境下的智能交通控制系统特点........................................................16
§3.6 本章小结....................................................................................................16
第四章 云环境下的智能交通任务调度...............................................................18
§4.1 概述............................................................................................................18
§4.2 云环境下的并行调度模型........................................................................18
§4.2.1 模型定义.............................................................................................18
§4.2.2 调度模型基本原理.............................................................................19
§4.3 基于改进蚁群优化算法的任务调度........................................................19
§4.3.1 改进蚁群算法的设计.........................................................................20
§4.3.2 改进蚁群算法的实现步骤.................................................................21
§4.4 仿真实验结果与分析................................................................................22
§4.5 本章小结....................................................................................................25
第五章 基于云模型的短时交通流预测...............................................................26
§5.1 交通流预测的基本参数............................................................................26
§5.2 交通流预测及优化模型——云模型........................................................26
§5.3 云模型下的短时交通流预测算法............................................................28
§5.3.1 云模型下的短时交通流预测原理.....................................................28
§5.3.2 交通流历史云模型的建立.................................................................28
§5.3.3 交通流预测云模型的建立.................................................................29
§5.4 云模型下的短时交通流预测仿真............................................................30
§5.5 仿真结果分析............................................................................................33

§5.6 本章小结....................................................................................................34
第六章 结论与展望...............................................................................................35
§6.1 总结............................................................................................................35
§6.2 展望............................................................................................................36
参考文献................................................................................................................37

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

15 积分 4人已下载

立即下载 VIP免费下载

摘要：

摘要在当今社会的交通领域，交通拥堵、交通安全以及交通污染问题已成为困扰人们出行过程中的三大难题，随着城市化进程速度的加快，汽车数量的增加速度已与道路里程、其他道路交通设施的增加速度严重失衡，从而导致了交通问题越来越严重。如何对海量交通信息进行高效处理、预测、调度、发布，实现人、车、路之间的信息共享、协同合作，减少交通拥挤和交通事故，提高交通效率，将是关系到未来交通信息服务的关键性问题。而云计算技术具有海量信息存储、计算资源统一管理、并行任务调度等优势，它能够有效地缓解道路交通信息系统所面临的诸多问题，实时的向出行者们发布动态交通服务信息、提供实时路况动态、预估出行时间、诱导行车路线等。因此，本...

展开>> 收起<<

云环境下智能交通控制系统的研究.doc

共41页,预览5页

还剩页未读，继续阅读

云环境下智能交通控制系统的研究

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

推荐作者

热门标签

举报选择: