基于D-S證據(jù)理論的城市交通流信息的融合方法

車(chē)輛工程專(zhuān)業(yè)畢業(yè)論文，論文編號(hào):JX913  論文字?jǐn)?shù):20272.頁(yè)數(shù):47

摘  要
 
 為了解決日益加劇的交通問(wèn)題困擾，產(chǎn)生了把交通基礎(chǔ)設(shè)施、交通運(yùn)載工具和交通參與者綜合起來(lái)系統(tǒng)考慮、充分利用高新技術(shù)解決交通問(wèn)題的智能交通系統(tǒng)。本文研究對(duì)象是智能交通系統(tǒng)中的重要主要組成部分--基于D-S證據(jù)理論的城市交通流信息的融合方法。
 城市交通中現(xiàn)有的交通流檢測(cè)方法比較豐富，但各個(gè)系統(tǒng)之間的交互性較差，獲取的交通流參數(shù)較單一；并且，由于檢測(cè)方法原理上和檢測(cè)設(shè)備性能上的不同，造成即使對(duì)同一組交通流參數(shù)采用不同的方法進(jìn)行檢測(cè)也會(huì)得到不同的結(jié)果；其三，缺乏有效的技術(shù)分析手段對(duì)這些數(shù)據(jù)進(jìn)行深層次挖掘，導(dǎo)致許多有重要價(jià)值的信息被湮沒(méi)和被忽略。針對(duì)以上問(wèn)題，本文提出基于多源交通流參數(shù)的證據(jù)合成方法，對(duì)目前主流的多種交通流傳感器采集的數(shù)據(jù)進(jìn)行分析、融合。
 城市交通流信息融合中的基于模糊粗糙集的證據(jù)理論研究與應(yīng)用：針對(duì)多種檢測(cè)方法（感應(yīng)線(xiàn)圈、視頻檢測(cè)、動(dòng)態(tài)OD 分析法和浮動(dòng)車(chē)法等）采集到的多源交通流數(shù)據(jù)，利用證據(jù)理論對(duì)其進(jìn)行融合處理，對(duì)融合過(guò)程中存在的檢測(cè)參數(shù)缺失、數(shù)據(jù)冗余，分別利用Greenshields交通流理論和改進(jìn)的粗糙集屬性約簡(jiǎn)規(guī)則進(jìn)行處理；在處理過(guò)程中采用模糊似然測(cè)度法獲取基本概率函數(shù)，并提出改進(jìn)的證據(jù)合成公式用以消解矛盾證據(jù)中的沖突。
 
關(guān)鍵詞：信息融合  動(dòng)態(tài)OD分析  證據(jù)理論  多源交通流數(shù)據(jù)    

ABSTRACT
   To address the growing problems of traffic, resulting in the transport infrastructure, transport vehicles and transport systems to consider together the participants make full use of high-tech solution to the traffic Intelligent Transportation System.This study is important in intelligent transport system main components based on D--S evidence theory of urban traffic flow information fusion method.
    Urban Transport in the existing traffic flow detection rich, but the interaction between the various systems is poor, access to the traffic flow over a single parameter; and, since the detection principle and the different testing equipment performance, resulting in eventhe same set of traffic flow parameters using different methods for detection will also get different results; Third, the lack of effective means of technical analysis of these data, deep excavation, leading to many important and valuable information is buried is ignored.To solve the above problem, this paper proposes multi-source traffic flow parameters based on the evidence synthesis, a variety of current mainstream traffic analysis of data collected by sensors, fusion.
  Urban traffic flow information fusion based on fuzzy rough set theory and application of evidence: for a variety of detection methods (induction coil, video detection, the first part of the dynamic OD analysis and floating car method, etc.) collected multiple sources of traffic flowdata, their use of evidence theory for fusion, the fusion process in the presence of the test parameters missing, data redundancy, respectively, using Greenshields improved traffic flow theory and rough set attribute reduction rules processing; in the process of the fuzzy similarHowever, access to basic probability function measurement method, and propose an improved formula for digestion and synthesis of evidence contradictory evidence in the conflict.
 
Key Words:   Information Fusion  Analysis of Dynamic OD  Evidence theory  Multi-source traffic flow data

目  錄
中文摘要.I
英文摘要.II
目    錄.IV
第一章  緒論.1
1.1  背景資料.1
1.2  研究主要內(nèi)容4
 1.2.1 交通流數(shù)據(jù)采集處理技術(shù)..4
 1.2.2 交通流數(shù)據(jù)挖掘與融合技術(shù). .5
1.3  研究主要目的7
 1.4  存在的問(wèn)題..7
第二章  城市交通的分析方法..9
2.1  路段行程時(shí)間的研究現(xiàn)狀..9
 2.2  滑動(dòng)窗口技術(shù)10
 2.3  典型行車(chē)路線(xiàn)的提出11
 2.4  系統(tǒng)設(shè)計(jì)模式12
 信息融合14
     3.1 信息融合技術(shù).14
     3.2 相關(guān)理論基礎(chǔ).15
 3.2.1 粗糙集理論..15
 3.2.2 粗糙集理論..16
 3.3 交通流信息融合模塊.17
第四章  城市交通流的信息融合過(guò)程..19
 4.1 交通流數(shù)據(jù)的預(yù)處理..19
   4.1.1 交通流檢測(cè)參數(shù)的插補(bǔ)..20
   4.1.2 利用粗糙集原理對(duì)原始數(shù)據(jù)進(jìn)行去冗余21
   4.1.3 交通流數(shù)據(jù)中屬性冗余性的消除22
 4.2 基于模糊集理論的基本概率函數(shù)求值23
 4.3 改進(jìn)的新的合成公式以及關(guān)聯(lián)交通參數(shù)的融合.24
第五章  實(shí)例分析與驗(yàn)證28
    5.1 原始數(shù)據(jù)的去冗余.30
    5.2 交通流參數(shù)的信息融合31
    5.3 數(shù)據(jù)驗(yàn)證34
結(jié)論36
參考文獻(xiàn)..37
附件40
致謝42