本文關(guān)鍵詞： 微觀交通流 數(shù)據(jù)驅(qū)動(dòng) 機(jī)器學(xué)習(xí) 跟馳模型 換道模型 NGSIM數(shù)據(jù)集　出處：《西南交通大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：跟馳模型和換道模型在交通安全評價(jià)、微觀交通仿真、自巡航控制、自動(dòng)駕駛等領(lǐng)域均有廣泛的應(yīng)用價(jià)值,它們也是道路微觀交通流理論的核心內(nèi)容。傳統(tǒng)的微觀交通流模型有一個(gè)共同不足就是它們均是基于數(shù)學(xué)公式和交通流理論建立的數(shù)學(xué)模型,這就導(dǎo)致此類模型很難有效地反映駕駛員的感知、思考、決策等一系列心理和生理活動(dòng)的不一致性和不確定性。本文從數(shù)據(jù)本身的角度出發(fā),利用機(jī)器學(xué)習(xí)相關(guān)算法特有優(yōu)勢,開展了基于數(shù)據(jù)驅(qū)動(dòng)的微觀交通流建模研究,來彌補(bǔ)上述不足,探索微觀交通流模型研究的新方向。首先,針對跟馳行為展開研究,以線性組合預(yù)測為基礎(chǔ),融合基于動(dòng)力學(xué)的跟馳模型對安全因素的可控性優(yōu)點(diǎn)和基于機(jī)器學(xué)習(xí)的跟馳模型的強(qiáng)大自學(xué)習(xí)優(yōu)點(diǎn),通過改進(jìn)最優(yōu)加權(quán)法中的目標(biāo)函數(shù),來建立線性組合跟馳模型;然后,針對換道行為中的換道決策階段,分別利用機(jī)器學(xué)習(xí)算法—BP神經(jīng)網(wǎng)絡(luò)、支持向量機(jī)、隨機(jī)森林建立了數(shù)據(jù)驅(qū)動(dòng)的自由換道決策模型,并利用歸一化、主成分分析法對NGSIM數(shù)據(jù)進(jìn)行預(yù)處理,然后用處理后的數(shù)據(jù)對模型進(jìn)行訓(xùn)練和測試,驗(yàn)證模型的有效性。另外,基于隨機(jī)森林獨(dú)有的優(yōu)勢對影響換道決策行為的因素的重要性進(jìn)行了分析;最后,針對換道行為中的換道執(zhí)行階段展開研究,首次提出基于BP神經(jīng)網(wǎng)絡(luò)建立換道執(zhí)行模型,借助BP神經(jīng)網(wǎng)絡(luò)強(qiáng)大的自學(xué)能力和非線性擬合能力等優(yōu)點(diǎn),來彌補(bǔ)傳統(tǒng)模型的不足。結(jié)果表明:組合跟馳模型的預(yù)測精度優(yōu)于Gipps模型,且可通過調(diào)整組合跟馳模型考慮真實(shí)性和安全性的權(quán)重,來達(dá)到控制預(yù)測速度的真實(shí)性和安全性的目的;數(shù)據(jù)驅(qū)動(dòng)的換道決策模型均有較高的精度;在換道決策階段,駕駛員更多地關(guān)注在于本車與目標(biāo)車道上車輛之間的距離,而不是相對速度,且相對目標(biāo)車道上的前車來說,后車對駕駛員的影響更大一些;數(shù)據(jù)驅(qū)動(dòng)的換道執(zhí)行模型具有非常高的精度,用數(shù)據(jù)驅(qū)動(dòng)的方法來建立換道執(zhí)行模型是可行且有效的。
[Abstract]:The following model and the change model have extensive application value in the fields of traffic safety evaluation, microscopic traffic simulation, self-cruise control, autopilot and so on. They are also the core contents of the road microscopic traffic flow theory. A common shortcoming of the traditional microscopic traffic flow models is that they are all mathematical models based on mathematical formulas and traffic flow theories. This makes it difficult for such models to effectively reflect the inconsistency and uncertainty of a series of psychological and physical activities such as perception, thinking, decision making and so on. From the point of view of data itself, this paper makes use of the unique advantages of machine learning related algorithms. In order to make up for the above deficiencies, a data-driven research on microscopic traffic flow modeling is carried out to explore the new direction of microscopic traffic flow model. Firstly, the research on car-following behavior is carried out, which is based on linear combination prediction. Combining the controllability of the dynamic car-following model to the security factors and the powerful self-learning advantage of the machine-learning-based car-following model, a linear combinatorial car-following model is established by improving the objective function in the optimal weighting method. In this paper, a data-driven decision model is established by using machine learning algorithm (-BP) neural network, support vector machine (SVM) and random forest, and the decision model is normalized. The principal component analysis (PCA) is used to preprocess the NGSIM data, and then the model is trained and tested with the processed data to verify the validity of the model. Based on the unique advantages of random forests, the importance of factors influencing the decision making behavior is analyzed. Finally, a new model based on BP neural network is proposed for the first time. The advantages of BP neural network such as self-learning ability and nonlinear fitting ability are used to make up for the shortcomings of the traditional models. The results show that the prediction accuracy of the combined car-following model is better than that of the Gipps model. We can control the reliability and safety of prediction speed by adjusting the weight of combination and car-following model to control the authenticity and security of prediction speed. The driver pays more attention to the distance between the vehicle and the vehicle in the target lane rather than the relative speed, and the rear car has more influence on the driver than the front car in the target lane. The data driven switch execution model has a high precision, and it is feasible and effective to establish the data driven method.
【學(xué)位授予單位】：西南交通大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：U491.112

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 邱小平;劉亞龍;馬麗娜;楊達(dá);;基于貝葉斯網(wǎng)絡(luò)的車輛換道模型[J];交通運(yùn)輸系統(tǒng)工程與信息;2015年05期

2 邱小平;劉亞龍;;基于支持向量機(jī)的車輛跟馳模型[J];重慶交通大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年06期

3 張凱;盧鄒穎;;短時(shí)車流量組合預(yù)測模型[J];南京信息工程大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年05期

4 錢超;許宏科;徐娜;;高速公路交通量組合預(yù)測模型研究[J];計(jì)算機(jī)仿真;2013年04期

5 楊達(dá);蒲云;楊飛;祝儷菱;;基于最優(yōu)間距的車輛跟馳模型及其特性[J];西南交通大學(xué)學(xué)報(bào);2012年05期

6 王建;鄧衛(wèi);趙金寶;;基于改進(jìn)型貝葉斯組合模型的短時(shí)交通流量預(yù)測[J];東南大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年01期

7 唐少先;黃志輝;陽春華;;基于模糊神經(jīng)網(wǎng)絡(luò)的車輛跟馳決策的研究[J];微計(jì)算機(jī)信息;2009年23期

8 周立軍;王殿海;李衛(wèi)青;;人工神經(jīng)網(wǎng)絡(luò)及粒子群優(yōu)化算法在跟馳模型中的應(yīng)用[J];吉林大學(xué)學(xué)報(bào)(工學(xué)版);2009年04期

9 張磊;李升波;王建強(qiáng);李克強(qiáng);;基于神經(jīng)網(wǎng)絡(luò)方法的集成式駕駛員跟車模型[J];清華大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年11期

10 李德慧;劉小明;榮建;胡江碧;;基于模糊神經(jīng)網(wǎng)絡(luò)的車輛跟馳建模與仿真研究[J];北京工業(yè)大學(xué)學(xué)報(bào);2007年04期

，

本文編號(hào)：1533885