面向高速公路行車安全預(yù)警的車道偏離及換道模型研究
發(fā)布時間：2018-07-14 22:04
【摘要】：隨著經(jīng)濟(jì)的快速發(fā)展，汽車保有量的迅猛增長，城市交通問題不斷加劇，主要表現(xiàn)為道路交通事故數(shù)量增加、交通擁堵以及相關(guān)的環(huán)境污染嚴(yán)重等。尤其是隨著高速公路建設(shè)水平的不斷提高，汽車行駛速度的不斷提高，交通事故數(shù)和傷亡人數(shù)逐年增長，交通安全越來越受到廣泛關(guān)注，已經(jīng)成為了全世界所面臨的重大問題。因此，展開對輔助駕駛員安全行駛的相關(guān)技術(shù)的研究，對于改善道路交通安全狀況具有十分重要的理論意義和實(shí)際應(yīng)用價值。 本文在對車輛安全輔助駕駛技術(shù)的發(fā)展及研究現(xiàn)狀回顧與分析的基礎(chǔ)上，選取對高速公路上車輛橫向駕駛行為的安全性進(jìn)行研究，主要研究內(nèi)容包括車輛車道偏離預(yù)警系統(tǒng)和車輛安全換道模型兩大部分，對其中關(guān)鍵理論與技術(shù)進(jìn)行了相關(guān)的研究與探討，完成的主要科研工作與取得的研究成果概括如下： 1、建立了車輛車道偏離時間TLC的估算模型。以車輛行駛過程中方向盤轉(zhuǎn)角是否零，將車輛的行駛軌跡劃分成直線和曲線軌跡兩種情況；同時，結(jié)合車輛運(yùn)動學(xué)理論、幾何知識和三角函數(shù)知識，對車輛分別行駛在直線和曲線路段上的實(shí)際軌跡進(jìn)行了研究，得到了車道偏離時間TLC的計(jì)算方法，由此建立了車輛向左和向右兩種偏離情況下偏離時間TLC的估算模型。分別就車輛橫向距離、相對偏航角和曲線道路半徑等參數(shù)對偏離時間TLC的影響進(jìn)行了數(shù)值仿真分析，得到了相應(yīng)的仿真結(jié)果；利用自主開發(fā)的車輛換道模擬系統(tǒng)進(jìn)行仿真實(shí)驗(yàn)，，得到了不同計(jì)算方法的TLC預(yù)測值及與真實(shí)TLC值的差值。數(shù)值仿真結(jié)果對于駕駛員的安全駕駛和道路線形的設(shè)計(jì)有一定的指導(dǎo)作用；實(shí)驗(yàn)結(jié)果表明本文建立的模型計(jì)算得到的TLC值比其他相似方法所得值更接近真實(shí)值，誤差更小。 2、提出了一種對駕駛員的橫向駕駛行為進(jìn)行分類的模糊決策系統(tǒng)。通過對駕駛員的歷史駕駛數(shù)據(jù)的分析，結(jié)合車輛車道偏離的運(yùn)動特性，文中選用橫向速度和橫向位置作為評價駕駛員橫向駕駛風(fēng)格的主要因素，分別求取它們的標(biāo)準(zhǔn)方差，并作為決策系統(tǒng)的輸入變量。然后利用基于Mamdami inference推理的模糊決策方法對駕駛員駕駛風(fēng)格類型進(jìn)行了分類，從而為建立具有自適應(yīng)特性的車道偏離預(yù)警系統(tǒng)打下基礎(chǔ)。 3、建立了基于駕駛員駕駛風(fēng)格類型和實(shí)時駕駛情況的可變虛擬車道邊界偏離預(yù)警模型。結(jié)合駕駛員駕駛風(fēng)格的類型、駕駛過程中實(shí)時橫向位移、橫向速度三個參數(shù)，文中利用Mamdami inference推理的模糊決策方法給出可變虛擬車道邊界的調(diào)整方案，能夠適應(yīng)不同的駕駛員類型和實(shí)時變化的駕駛情況。然后，綜合分析車道偏離預(yù)警系統(tǒng)的特性和現(xiàn)有研究中采用的預(yù)警性能指標(biāo)，本文選用了虛警率和預(yù)警時間作為可變虛擬車道邊界偏離預(yù)警模型的評價指標(biāo)。最后，由于實(shí)驗(yàn)設(shè)備和條件有限，本文設(shè)置了虛擬實(shí)驗(yàn)仿真，詳細(xì)介紹了實(shí)驗(yàn)方法、評價方法以及跟其他方法的結(jié)果對比的表現(xiàn)形式。 4、建立了基于最小安全距離的車輛換道模型。首先，從車輛換道行為特性和運(yùn)動學(xué)特性的角度出發(fā)，詳細(xì)描述車輛換道過程，給出了換道的三階段和碰撞情況的分類，同時定義了車輛換道環(huán)境、運(yùn)動學(xué)參數(shù)和時間參數(shù)等。然后，從車輛換道的運(yùn)動學(xué)角度出發(fā)，分析了換道車輛M與其周圍相鄰車輛（車輛Mol、Mdl、Mdf）在換道過程中的運(yùn)動位置關(guān)系，以換道過程中避免碰撞為目標(biāo)，建立了換道車輛M與相關(guān)聯(lián)車輛之間的最小初始安全換道距離模型。 5、對換道車輛M以勻速和加速兩種換道策略進(jìn)行了仿真分析。以建立的最小安全距離換道模型為基礎(chǔ)，分別對換道車輛M設(shè)置了兩種不同的換道策略，分別為：一是車輛M以勻速狀態(tài)進(jìn)行換道；二是車輛M以加速狀態(tài)進(jìn)行換道。利用Matlab仿真軟件對兩種不同換道的情況進(jìn)行了仿真分析，給出了車輛M與周圍相鄰車輛間的安全換道區(qū)域和非安全換道區(qū)域。對仿真結(jié)果進(jìn)行了分析，得到了影響車輛安全換道的相關(guān)因素，同時，通過對比分析車輛M在勻速換道和加速換道情況下的安全區(qū)域范圍可知，車輛M采取加速度換道策略，可以拓寬安全區(qū)域范圍，換道安全更可靠和安全。
[Abstract]:With the rapid development of the economy and the rapid growth of the car ownership, the urban traffic problems are increasing, mainly as the increase in the number of road traffic accidents, traffic congestion and the serious environmental pollution, especially with the continuous improvement of the highway construction level, the continuous improvement of the speed of the car and the number of traffic accidents and casualties. As the number of people is increasing year by year, traffic safety has been paid more and more attention, and it has become a major problem in the world. Therefore, it is of great theoretical and practical value to improve the safety of road traffic.
On the basis of the review and analysis of the development and research status of vehicle safety assisted driving technology, this paper studies the safety of the lateral driving behavior of vehicles on the highway. The main research contents include two parts of the vehicle lane departure warning system and the vehicle safety change model, which are the key theories and technologies. The main research work and achievements obtained are summarized as follows:
1, the estimation model of vehicle lane departure time TLC is established. In the course of the vehicle, whether the steering angle of the steering wheel is zero is zero and the vehicle trajectory is divided into two cases of straight line and curve trajectory. At the same time, combined with the theory of vehicle kinematics, geometry knowledge and trigonometric function knowledge, the vehicle is running on the straight line and the curve section respectively. The trajectory is studied and the calculation method of lane departure time TLC is obtained. Therefore, the estimation model of departure time TLC is established under two deviations of left and right vehicles. The effects of vehicle lateral distance, relative yaw angle and curve road radius on the deviation time TLC are numerically simulated and analyzed, and the corresponding results are obtained. The simulation results of the simulation are carried out by the self developed vehicle switching simulation system. The TLC prediction value of different calculation methods and the difference with the real TLC value are obtained. The numerical simulation results have certain guiding role for the driver's safe driving and road alignment design; the experimental results show that the model established in this paper is calculated. The TLC value is closer to the real value than other similar methods, and the error is smaller.
2, a fuzzy decision system is put forward to classify the driver's lateral driving behavior. Through the analysis of the driver's historical driving data and the motion characteristics of the vehicle lane deviation, the lateral velocity and lateral position are selected as the main factors to evaluate the driver's lateral driving wind. It is the input variable of the decision system. Then, the type of driver's driving style is classified by using the fuzzy decision method based on Mamdami inference reasoning, which lays the foundation for the establishment of lane departure warning system with adaptive characteristics.
3, a variable virtual lane boundary departure warning model based on driver style type and real time driving condition is established. Combining the type of driver's driving style, real-time lateral displacement and lateral velocity of three parameters during driving, the fuzzy decision method of Mamdami inference reasoning is used to give the adjustment of the variable virtual lane boundary. The whole scheme can adapt to different driver types and real-time changing driving conditions. Then, the characteristics of the lane departure warning system and the early warning performance indicators used in the existing research are synthetically analyzed. The false alarm rate and early warning time are selected as the evaluation indexes of the early warning model of the variable virtual lane boundary deviation. Finally, the experiment is made. The equipment and conditions are limited. In this paper, a virtual experiment simulation is set up, and the experimental method, evaluation method and the form of comparison with other methods are introduced in detail.
4, a vehicle switching model based on the minimum safe distance is established. First, the vehicle change process is described in detail from the point of view of the behavior and kinematics of the vehicle. The three stages of the change and the classification of the collision are given, and the vehicle changing environment, the kinematic parameters and the time parameters are defined. On the basis of the kinematic point of view, the kinematic position relationship between the M and its adjacent vehicles (Mol, Mdl, Mdf) in the changing course is analyzed. The minimum initial safety change distance model between the lane changing vehicle M and the associated vehicle is established by avoiding the collision in the course of changing the path.
5, the simulation analysis is carried out on the two lane changing strategy of the change vehicle M. Based on the minimum safe distance change model, two different switching strategies are set up for the change vehicle M respectively. The first is that the vehicle M is changing the path in a uniform state, and the two is the vehicle M changing the channel in an accelerated state. The Matlab imitation is used. The real software has simulated and analyzed the two different ways of changing the channel. The safety change area and the unsafe lane change area between the vehicle M and the adjacent vehicles are given. The simulation results are analyzed, and the related factors affecting the vehicle safety change are obtained. At the same time, the vehicle M is compared and analyzed in the condition of uniform speed change and acceleration change. The safe area range shows that vehicle M adopts acceleration lane changing strategy, which can widen the safety area and change the road safety more reliably and safely.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：U492.84;U491.25
本文編號：2123101
[Abstract]:With the rapid development of the economy and the rapid growth of the car ownership, the urban traffic problems are increasing, mainly as the increase in the number of road traffic accidents, traffic congestion and the serious environmental pollution, especially with the continuous improvement of the highway construction level, the continuous improvement of the speed of the car and the number of traffic accidents and casualties. As the number of people is increasing year by year, traffic safety has been paid more and more attention, and it has become a major problem in the world. Therefore, it is of great theoretical and practical value to improve the safety of road traffic.
On the basis of the review and analysis of the development and research status of vehicle safety assisted driving technology, this paper studies the safety of the lateral driving behavior of vehicles on the highway. The main research contents include two parts of the vehicle lane departure warning system and the vehicle safety change model, which are the key theories and technologies. The main research work and achievements obtained are summarized as follows:
1, the estimation model of vehicle lane departure time TLC is established. In the course of the vehicle, whether the steering angle of the steering wheel is zero is zero and the vehicle trajectory is divided into two cases of straight line and curve trajectory. At the same time, combined with the theory of vehicle kinematics, geometry knowledge and trigonometric function knowledge, the vehicle is running on the straight line and the curve section respectively. The trajectory is studied and the calculation method of lane departure time TLC is obtained. Therefore, the estimation model of departure time TLC is established under two deviations of left and right vehicles. The effects of vehicle lateral distance, relative yaw angle and curve road radius on the deviation time TLC are numerically simulated and analyzed, and the corresponding results are obtained. The simulation results of the simulation are carried out by the self developed vehicle switching simulation system. The TLC prediction value of different calculation methods and the difference with the real TLC value are obtained. The numerical simulation results have certain guiding role for the driver's safe driving and road alignment design; the experimental results show that the model established in this paper is calculated. The TLC value is closer to the real value than other similar methods, and the error is smaller.
2, a fuzzy decision system is put forward to classify the driver's lateral driving behavior. Through the analysis of the driver's historical driving data and the motion characteristics of the vehicle lane deviation, the lateral velocity and lateral position are selected as the main factors to evaluate the driver's lateral driving wind. It is the input variable of the decision system. Then, the type of driver's driving style is classified by using the fuzzy decision method based on Mamdami inference reasoning, which lays the foundation for the establishment of lane departure warning system with adaptive characteristics.
3, a variable virtual lane boundary departure warning model based on driver style type and real time driving condition is established. Combining the type of driver's driving style, real-time lateral displacement and lateral velocity of three parameters during driving, the fuzzy decision method of Mamdami inference reasoning is used to give the adjustment of the variable virtual lane boundary. The whole scheme can adapt to different driver types and real-time changing driving conditions. Then, the characteristics of the lane departure warning system and the early warning performance indicators used in the existing research are synthetically analyzed. The false alarm rate and early warning time are selected as the evaluation indexes of the early warning model of the variable virtual lane boundary deviation. Finally, the experiment is made. The equipment and conditions are limited. In this paper, a virtual experiment simulation is set up, and the experimental method, evaluation method and the form of comparison with other methods are introduced in detail.
4, a vehicle switching model based on the minimum safe distance is established. First, the vehicle change process is described in detail from the point of view of the behavior and kinematics of the vehicle. The three stages of the change and the classification of the collision are given, and the vehicle changing environment, the kinematic parameters and the time parameters are defined. On the basis of the kinematic point of view, the kinematic position relationship between the M and its adjacent vehicles (Mol, Mdl, Mdf) in the changing course is analyzed. The minimum initial safety change distance model between the lane changing vehicle M and the associated vehicle is established by avoiding the collision in the course of changing the path.
5, the simulation analysis is carried out on the two lane changing strategy of the change vehicle M. Based on the minimum safe distance change model, two different switching strategies are set up for the change vehicle M respectively. The first is that the vehicle M is changing the path in a uniform state, and the two is the vehicle M changing the channel in an accelerated state. The Matlab imitation is used. The real software has simulated and analyzed the two different ways of changing the channel. The safety change area and the unsafe lane change area between the vehicle M and the adjacent vehicles are given. The simulation results are analyzed, and the related factors affecting the vehicle safety change are obtained. At the same time, the vehicle M is compared and analyzed in the condition of uniform speed change and acceleration change. The safe area range shows that vehicle M adopts acceleration lane changing strategy, which can widen the safety area and change the road safety more reliably and safely.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2014
【分類號】：U492.84;U491.25
【參考文獻(xiàn)】
相關(guān)期刊論文 前10條
1 常華耀;王軍政;陳超;李靜;;一種基于光照無關(guān)圖的車道檢測方法[J];北京理工大學(xué)學(xué)報;2011年11期
2 時柏營;楊曉光;于曉斐;;考慮與跟馳模型結(jié)合的自由換道模型及仿真[J];交通科學(xué)與工程;2009年04期
3 劉潔;黃斌;;單目視覺車道偏離報警系統(tǒng)[J];電子產(chǎn)品世界;2007年03期
4 廖傳錦,黃席樾,柴毅,汪先矩,劉衛(wèi)紅;基于邊緣提取和特征跟蹤的道路檢測算法[J];重慶大學(xué)學(xué)報(自然科學(xué)版);2002年01期
5 王家凡,羅大庸;交通流微觀仿真中的換道模型[J];系統(tǒng)工程;2004年03期
6 莊勁松;于鵬輝;曲大義;陳秀峰;;高速公路交通安全預(yù)警系統(tǒng)的研究[J];公路交通科技(應(yīng)用技術(shù)版);2011年03期
7 王榮本,李斌,儲江偉,郭克友;世界智能車輛行駛安全保障技術(shù)的研究進(jìn)展[J];公路交通科技;2002年02期
8 楊建國,王金梅,李慶豐,王兆安;微觀仿真中車輛換道的行為分析和建模[J];公路交通科技;2004年11期
9 李瑋;高德芝;段建民;;智能車輛自由換道模型研究[J];公路交通科技;2010年02期
10 許倫輝;羅強(qiáng);吳建偉;黃艷國;;基于最小安全距離的車輛跟馳模型研究[J];公路交通科技;2010年10期
相關(guān)博士學(xué)位論文 前2條
1 陳軍;基于DSP的高速公路車道偏離報警系統(tǒng)研究[D];天津大學(xué);2010年
2 陳曉冬;基于交通流理論的高速公路安全預(yù)警系統(tǒng)關(guān)鍵技術(shù)研究[D];吉林大學(xué);2011年
本文編號：2123101
本文鏈接：http://sikaile.net/kejilunwen/jiaotonggongchenglunwen/2123101.html
最近更新
教材專著
