本文選題：路徑誘導(dǎo) + 線性回歸��； 參考：《北京郵電大學(xué)》2014年碩士論文

【摘要】：隨著城市的發(fā)展,交通擁堵問題越來越嚴(yán)重,交通擁堵可導(dǎo)致車輛通行速度下降,并因此帶來汽車尾氣污染等問題,影響了人民的生活。路徑誘導(dǎo)技術(shù)是提高道路的通行效率,避免交通擁堵的有效方法,同時也是智能交通領(lǐng)域中的一個重要組成部分。準(zhǔn)確可靠的交通流信息是路徑誘導(dǎo)系統(tǒng)中的關(guān)鍵數(shù)據(jù)和基礎(chǔ)數(shù)據(jù)。本文在前人研究的基礎(chǔ)上對智能交通領(lǐng)域的短時交通流預(yù)測和路徑誘導(dǎo)兩個方面進(jìn)行分析和研究。本文的主要工作如下： 1)設(shè)計了仿真實驗,用于驗證基于道路拓?fù)浜蜋?quán)重的地圖匹配算法的有效性,該實驗?zāi)M真實路況產(chǎn)生的流程,使用真實車輛的GPS數(shù)據(jù),得到車輛通過道路的平均速度。 2)設(shè)計并實現(xiàn)了基于線性回歸的短時交通流預(yù)測算法,使用道路速度的歷史數(shù)據(jù),在算法的選擇上,綜合考慮線性回歸,神經(jīng)網(wǎng)絡(luò),支持向量機(jī),選用更適合于工程環(huán)境實現(xiàn)的線性回歸算法,設(shè)計線性回歸模型的輸入?yún)?shù),使之能有效預(yù)測道路的短時速度。實現(xiàn)一個改進(jìn)的拓?fù)渑判虻恼{(diào)度算法,減少算法在訓(xùn)練時的內(nèi)存使用。設(shè)計實驗,驗證了算法預(yù)測的準(zhǔn)確性。 3)設(shè)計了基于動態(tài)交通信息的路徑誘導(dǎo)算法,根據(jù)道路長度,實時速度和短時預(yù)測速度設(shè)計了一個動態(tài)的權(quán)重計算方法,使路徑誘導(dǎo)算法能有效規(guī)避擁堵的道路。設(shè)計了仿真實驗進(jìn)行驗證,表明算法具有一定的有效性。
[Abstract]:With the development of the city, the problem of traffic congestion becomes more and more serious. Traffic congestion can lead to the decrease of the speed of traffic and the pollution of vehicle exhaust, which affects people's life. Route guidance technology is an effective method to improve the traffic efficiency and avoid traffic congestion. It is also an important part of intelligent transportation field. Accurate and reliable traffic flow information is the key data and basic data in the path guidance system. On the basis of previous studies, this paper analyzes and studies the two aspects of short time traffic flow prediction and route guidance in the field of intelligent transportation. The main work of this paper is as follows: 1) A simulation experiment is designed to verify the validity of the map matching algorithm based on the road topology and weight. The experiment simulates the flow of the real road conditions and uses the GPS data of the real vehicle to get the average speed of the vehicle passing through the road. 2) the short-term traffic flow prediction algorithm based on linear regression is designed and implemented. Using the historical data of road speed, the linear regression, neural network and support vector machine are considered in the selection of the algorithm. The linear regression algorithm which is more suitable for engineering environment is selected and the input parameters of the linear regression model are designed so that it can effectively predict the short time speed of the road. An improved scheduling algorithm for topological sorting is implemented to reduce the memory usage of the algorithm in training. Experiments are designed to verify the accuracy of the algorithm. 3) A path guidance algorithm based on dynamic traffic information is designed, and a dynamic weight calculation method is designed according to road length, real time speed and short time prediction speed, which can effectively avoid congested roads. Simulation experiments are designed to verify the effectiveness of the algorithm.
【學(xué)位授予單位】：北京郵電大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：U495;TP301.6

【參考文獻(xiàn)】

相關(guān)期刊論文 前9條

1 彭飛,柳重堪,張其善;基于代價函數(shù)的組合導(dǎo)航系統(tǒng)地圖匹配算法[J];北京航空航天大學(xué)學(xué)報;2002年03期

2 田晶;楊玉珍;陳陽舟;;短時交通流量兩種預(yù)測方法的研究[J];公路交通科技;2006年04期

3 李兵;鄭四發(fā);曹劍東;連小珉;李克強(qiáng);;基于路網(wǎng)分層策略的多源點最短距離算法[J];計算機(jī)工程;2008年12期

4 劉杰;黃亞樓;;基于BP神經(jīng)網(wǎng)絡(luò)的非線性網(wǎng)絡(luò)流量預(yù)測[J];計算機(jī)應(yīng)用;2007年07期

5 史其信,鄭為中;道路網(wǎng)短期交通流預(yù)測方法比較[J];交通運輸工程學(xué)報;2004年04期

6 黃仝宇;胡剛;傅惠;;基于增量搜索的車輛動態(tài)路徑規(guī)劃算法研究[J];西部交通科技;2010年09期

7 段凡丁;關(guān)于最短路徑的SPFA快速算法[J];西南交通大學(xué)學(xué)報;1994年02期

8 章威;徐建閩;林綿峰;;基于大規(guī)模浮動車數(shù)據(jù)的地圖匹配算法[J];交通運輸系統(tǒng)工程與信息;2007年02期

9 張學(xué)敏;張航;;基于改進(jìn)蟻群算法的最短路徑問題研究[J];自動化技術(shù)與應(yīng)用;2009年06期

，

本文編號：1844408