基于ANSYS的汽車驅(qū)動橋殼的有限元分析

基于ANSYS的汽車驅(qū)動橋殼的有限元分析,基于,ansys,汽車,驅(qū)動,有限元分析 畢 業(yè) 設(shè) 計（論 文）外 文 參 考 資 料 及 譯 文 譯文題目： Autonomous Intelligent Vehicles 自動駕駛智能汽車 學(xué)生姓名： ?！　I(yè)： 所在學(xué)院： 指導(dǎo)教師： 職　　稱： 說明： 要求學(xué)生結(jié)合畢業(yè)設(shè)計（論文）課題參閱一篇以上的外文資料，并翻譯至少一萬印刷符（或譯出3千漢字）以上的譯文。譯文原則上要求打印（如手寫，一律用400字方格稿紙書寫），連同學(xué)校提供的統(tǒng)一封面及英文原文裝訂，于畢業(yè)設(shè)計（論文）工作開始后2周內(nèi)完成，作為成績考核的一部分。 Autonomous Intelligent Vehicles 1.1 Research Motivation and Purpose Autonomous intelligent vehicles are generic technology sets to augment vehicle autonomous driving entirely or in part for autonomous and safety purposes. Fundamentally, autonomous intelligent vehicles refer to many mobile robot technologies. In principle, we consider autonomous intelligent vehicles as mobile robot platforms in this book. Hence, an intelligent vehicle consists of four fundamental technologies: environment perception and modeling, localization and map building, path planning and decision-making, and motion control [26], shown in Fig. 1.1. The dreams of a human being are the power and source of pushing the world forward. The National Research Council once predicted that the core weapon in the twentieth century would be the tank, while that in the twenty-first century—an unmanned battle system [1]. Moreover, a third of the U.S. military ground vehicles must be unmanned by 2015. Therefore, since 1980s, the Defense Advanced Research Projects Agency (DARPA) initiated a new project, namely the unmanned battle project. Its goal is to design a car which can autonomously implement navigation, obstacle avoidance, and path planning. Afterwards, it opened an intelligent vehicle era. Moreover, the U.S. Department of Energy launched a ten year robot and intelligent system plan (1986–1995), and also the space robot plan. In terms of space exploration, the National Aeronautics and Space Administration (NASA) has developed several wheeled rovers, such as Spirit and Opportunity, for science explorations.1 A major concern associated with the rapid growth in automotive production is an increase in traffic congestion and accidents [36]. To solve the problem, the governments all over the world have been increasing funds to improve the traffic infrastructure, enforce traffic laws, and educate drivers about traffic regulations. In addition, research institutes have launched R&D projects in driver assistance and safety warning systems. Therefore, in the last decade, many research works in the area of intelligent vehicles all over the world led to Intelligent Transportation Systems (ITS) for improving road safety and reducing traffic accidents [7]. Autonomous intelligent vehicles are now widely applied to Driver Assistance and Safety Warning Systems (DASWS) [36], such as Forward Collision Warning [9, 27], Adaptive Cruise Control [32], Lane Departure Warning [16]. In recent years, with the development of economy and society, the issues of traffic safety, energy shortage, and environment pollution became more serious. Those problems then led to higher volumes of research and applications. Toward this end, combining vehicles, drivers and lanes together, we can implement better traffic capacity and traffic safety using computer control, artificial intelligence and communication technologies [3]. The most important reasons for the large numbers of traffic accidents are burdensome driving and fatigue driving. When driving on the traffic congestion lanes, drivers have to do a lot of operations, such as shifting and pulling clutches, and they have to complete 20 to 30 coordination operations of hand and foot movements each minute. With the economic development and the increase of vehicle ownership, the number of non-professional drivers are rising, leading to frequent traffic accidents. As a result, traffic accidents have become the first public nuisance in modern society. Traffic problems have troubled the whole world, and then, the question of how to improve traffic safety has become an urgent social issue. Lane departure systems, fatigue detection systems, and automatic cruise control can greatly reduce driver’s workload and improve transportation system safety. The widely application prospects of intelligent vehicles promote the development of transportation systems which attracts a growing number of research institutions and auto manufacturers. The DARPA had held the Grand Challenges and the Ur- ban Challenge since 2004. Their goal is to develop autonomous intelligent vehicles capable of both perceiving various environments, such as desert trails, roads, and urban areas, and navigating at high speeds2 [5, 30, 31]. In the first Grand Challenge, CMU’s Sandstorm went for 7.4 miles from the start, opening the possibilities of autonomous capability [30]. In 2005, five vehicles, namely Stanley, Sand- storm, Highlander, Kat-5, and Terra-Max, were able to complete that challenge, and Stanley took the first place ahead of Sandstorm [31]. After the success of the two Grand Challenges, the DARPA organized the Urban Challenge [5]. In the Urban Challenge, based on the technical reports of implementing safe and capable autonomous vehicles, the DARPA allowed 53 teams to demonstrate how they navigate simple urban driving scenes. After these demonstrations, only 36 teams were invited to attend the National Qualification Event (NQE). Finally, only 11 teams were qualified for the Urban Challenge Final Event (UCFE). In China, the 2008 Beijing Olympic Games whose slogans were Hi-tech Olympics and Green Olympics adopted many advanced traffic management systems, intelligent vehicles, electric vehicles for improving vehicle safety performance, reducing pollution, easing traffic congestion. Consequently, those innovations drew attention of many researchers. In 2011, China released ten leading edge technologies and modern transportation technologies among which were technologies aiming at developing intelligent vehicles. Moreover, the National Natural Science Foundation of China launched the state key development plan in 2008, so that audio-visual information based cognizing computation3 could integrate human–computer interfaces, computer vision, language understanding, and cooperative computing. Finally, upon those achievements, the goal of this plan is to develop autonomous intelligent vehicles which are capable of both perceiving natural environment and making intelligent decisions. Meanwhile, similar to the Grand Challenge supported by DARPA, the plan holds the Future Challenge each year. The research on intelligent vehicles can greatly facilitate the rapid development of other disciplines, such as exploring planets. The U.S. Mars vehicles Spirit and Opportunity play an irreplaceable role in exploring Mars and the vast universe be- yond Mars [13, 23]. In China, the government released the White Paper “China Aerospace” in November 2000, which targets exploring the moon and other planets in the near future. Furthermore, space mobile robots are the key part for exploring planets which could benefit the utilizing solar energy. 1.2 The Key Technologies of Intelligent Vehicles As we mentioned before, intelligent vehicles are a set of intelligent agents which integrate multi-sensor fusion based environment perception and modeling, localization and map building, path planning and decision-making, and motion control, shown in Fig. 1.1. The environment perception and modeling module is responsible for sensing environment structures in a multi-sensor way and providing a model of the surrounding environment. Here, the environment model includes a list of moving objects，that of static obstacles, vehicle position relative to the current road, the road shape, etc. Finally, this module provides the environment model and the local map to the localization and map building module by processing the original data, vision, lidar, and radar. The second module, vehicle localization and map building, is to use geometric feature location estimate in the map to determine the vehicle’s position, and to interpret sensor information to estimate the locations of geometric features in a global map. As a result, the second module yields a global map based on the environment model and a local map. The path planning and decision-making module is to assist in ensuring that the vehicle is operated in accordance with the rules of the ground, safety, comfortability, vehicle dynamics, and environment con- texts. Hence, this module can potentially improve mission efficiency and generate the desired path. The final module, motion control, is to execute the commands necessary to achieve the planned paths, thus yielding interaction between the vehicle and its surrounding environment. A brief introduction of these modules is presented below. 1.2.1 Multi-sensor Fusion Based Environment Perception and Modeling Figure 1.2 illustrates a general environment perception and modeling framework. From this framework, we can see that: (i) The original data are collected by various sensors; (ii) Various features are extracted from the original data, such as road (object) colors, lane edges, building contours; (iii) Semantic objects are recognized using classifiers, and consist of lanes, signs, vehicles, pedestrians; (iv) We can de- duce driving contexts, and vehicle positions. Multi-sensor fusion Multi-sensor fusion is the basic framework of intelligent vehicles for better sensing surrounding environment structures, and detecting objects/obstacles. Roughly, the sensors used for surrounding environment perception are divided into two categories: active and passive ones. Active sensors include lidar, radar, ultrasonic and radio, while the commonly-used passive sensors are infrared and visual cameras. Different sensors are capable of providing different detection precision and range, and yielding different effects on environment. That is, combining various sensors could cover not only short-range but also long-range objects/obstacles, and also work in various weather conditions. Furthermore, the original data of different sensors can be fused in low-level fusion, high-level fusion, and hybrid fusion [4, 14, 20, 35]. Dynamic Environment Modeling Dynamic environment modeling based on moving on-vehicle cameras plays an important role in intelligent vehicles [17]. However, this is extremely challenging due to the combined effects of ego-motion, blur, light changing. Therefore，traditional methods for gradual illumination change, small motion objects [28] such as background subtraction, do not work well any more, even those that have been widely used in surveillance applications. Consequently, more and more approaches try to handle these issues [2, 17]. Unfortunately, it is still an open problem to reliably model and update background. To select different driving strategies, several broad scenarios are usually considered in path planning and decision-making, when navigating roads, intersections, parking lots, jammed intersections. Hence, scenario estimators are helpful for further decision-making, which is commonly used in the Urban Challenge. Object Detection and Tracking In general, in a driving environment, we are interested in static/dynamic obstacles, lane markings, traffic signs, vehicles, and pedestrians. Correspondingly, object detection and tracking are the key parts of environment perception and modeling. 1.2.2 Vehicle Localization and Map Building The goal of vehicle localization and map building is to generate a global map by combining the environment model, a local map and global information. In autonomous driving, vehicle localization is either to estimate road geometry or to localize the vehicle relative to roads under the conditions of known maps or un- known maps. Hence, vehicle localization refers to road shape estimation, position filtering, transforming the vehicle pose into a coordinate frame. For vehicle localization, we face several challenges as follows: (i) Usually, the absolute positions from GPS/DGPS and its variants are insufficient due to signal transmission; (ii) The path planning and decision-making module needs more than just the vehicle absolute position as input; (iii) Sensor noises greatly affect the accuracy of vehicle localization. Regarding the first issue, though the GPS and its variants have been widely used in vehicle localization, its performance could degrade due to signal blockages and reflections of buildings and trees. In the worst case, Inertia Navigation Sys- tem (INS) can maintain a position solution. As for the second issue, local maps fusing laser, radar, and vision data with vehicle states are used to locate and track both static/dynamic obstacles and lanes. Furthermore, global maps could contain lane geometric information, lane makings, step signs, parking lots, check points and provide global environment information. Referring to the third issue, various noise modules are considered to reduce localization error [26]. 自動駕駛智能汽車 1.1研究的初衷和目的 自主智能車是通用的技術(shù)設(shè)置，以增加車輛自主駕駛或部分自主駕駛的安全性為目的。從根本上說，自主的智能車輛是指多移動機(jī)器人技術(shù)。原則上，我們認(rèn)為自主智能車應(yīng)該放在移動機(jī)器人平臺這本書中。因此，智能車輛包括四個基本技術(shù)：環(huán)境感知和建模，地圖創(chuàng)建與定位，路徑規(guī)劃和決策，以及運動控制。[26] 如圖1.1所示。 一個人的夢想是推動世界前進(jìn)的動力和源泉。美國國家研究委員會曾預(yù)測，二十世紀(jì)的核心武器是坦克，而在二十一世紀(jì)是無人戰(zhàn)斗系統(tǒng)[ 1 ] 。此外到2015年,美國軍隊的地面無人車輛要達(dá)到總車輛的三分之一。因此，20世紀(jì)80年代以來，美國國防高級研究計劃局（DARPA ）啟動了一個新項目，即無人駕駛戰(zhàn)斗項目。它的目標(biāo)是設(shè)計一款能實現(xiàn)自主導(dǎo)航，避障，路徑規(guī)劃的車。此后，它開啟了一個智能汽車時代。此外，美國能源部推出一個十年機(jī)器人和智能系統(tǒng)規(guī)劃（ 1986-1995 ） ，也是太空機(jī)器人計劃。在太空探索方面，美國國家航空和航天局（NASA ）已經(jīng)開發(fā)了數(shù)個輪式探測器，如勇氣號和機(jī)遇號等。 隨著汽車產(chǎn)量的快速增長，出現(xiàn)了交通擁堵和交通事故等主要問題。[36]。為了解決這個問題，世界各地的政府一直在加大資金改善交通基礎(chǔ)設(shè)施，加強(qiáng)交通法規(guī)，和教給司機(jī)交通法規(guī)。此外，研究機(jī)構(gòu)開展了駕駛員輔助和安全預(yù)警系統(tǒng)研發(fā)項目。因此，在過去的十年中，世界各地在智能車系統(tǒng)（ITS）上開展研究工作，以改善道路安全，減少區(qū)域交通事故來實現(xiàn)智能交通。[7]。巡航智能車現(xiàn)在被廣泛應(yīng)用到駕駛輔助和安全預(yù)警系統(tǒng)（DASWS）[36]，如前部碰撞警告[9，27]，自適應(yīng)巡航控制系統(tǒng)[32]，車道偏離警告[16]。近年來，隨著社會經(jīng)濟(jì)的發(fā)展，交通安全，能源短缺和環(huán)境污染問題變得更加嚴(yán)重。這些問題則需要更高層次的研究。為實現(xiàn)這一目標(biāo)，把車輛，駕駛員和通道結(jié)合起來，我們就可以用電腦控制，人工智能和通訊技術(shù)才能實現(xiàn)更好的通行能力和交通安全[3]。 發(fā)生大量交通事故的最主要的原因是繁重的駕駛和疲勞駕駛。當(dāng)交通擁堵時，司機(jī)必須做很多操作，比如踩離合器和拉手剎，他們必須每分鐘完成20至30次的協(xié)調(diào)動作。隨著經(jīng)濟(jì)的發(fā)展和汽車擁有量的增加，非職業(yè)司機(jī)的人數(shù)正在上升，導(dǎo)致交通事故頻發(fā)。這樣一來，交通事故已經(jīng)成為現(xiàn)代社會第一公害。交通問題已經(jīng)困擾了整個世界，那么，如何提高行車安全已成為一個緊迫的社會問題。車道偏離系統(tǒng)，疲勞檢測系統(tǒng)，自動巡航控制系統(tǒng)可大大降低駕駛員的勞動強(qiáng)度，提高交通系統(tǒng)的安全性。 智能汽車應(yīng)用前景廣泛，促進(jìn)了交通運輸系統(tǒng)的發(fā)展，吸引了越來越多的研究機(jī)構(gòu)和汽車制造商。從2004年開始，美國國防高級研究計劃局開始舉辦智能汽車大挑戰(zhàn)和城市挑戰(zhàn)賽。他們的目標(biāo)是開發(fā)自主智能車輛在高速行駛的情況下能夠感知各種環(huán)境的功能，如沙漠小徑，道路和城市地區(qū)。[5，30，31].在第一次大挑戰(zhàn)中CMU的沙塵暴從一開始行駛到7.4英里處開始實現(xiàn)自主智能。[30].2005年有5輛汽車能夠完成這一挑戰(zhàn)，分別是斯坦利、沙塵暴、漢蘭達(dá)、 kat-5和TerraMax。斯坦利取代了沙塵暴第一的位置。[31]. 這兩個大挑戰(zhàn)成功后，DARPA舉辦了城市挑戰(zhàn)賽。在城市挑戰(zhàn)賽中，基于車輛自主安全的技術(shù)檢測，DARPA允許53支車隊展示他們是如何在城市中簡單駕駛的場景。在這些展示之后，只有36支車隊有資格被邀請參加全國性的賽事。最后，只有11支車隊有資格進(jìn)入城市挑戰(zhàn)賽的決賽。在中國，2008年北京奧運會的宗旨是科技奧運、綠色奧運，采用了許多先進(jìn)的交通管理系統(tǒng)和電動汽車，來提高車輛的安全性能，減少污染，緩解交通堵塞。因此，這些創(chuàng)新吸引了眾多研究者的關(guān)注。2011，中國發(fā)布了十大前沿技術(shù)和現(xiàn)代交通技術(shù)，其中以開發(fā)智能車技術(shù)為目標(biāo)。此外，2008年，中國國家自然科學(xué)基金推出了國家重點發(fā)展規(guī)劃，以視聽信息為基礎(chǔ) 使認(rèn)知計算來整合人機(jī)界面、計算機(jī)視覺、語言理解、協(xié)同計算。最后，該計劃的目標(biāo)是開發(fā)能夠自主認(rèn)知自然環(huán)境和智能決策的智能車輛，直到這些計劃能夠達(dá)到要求。同時，類似于DARPA支持的大挑戰(zhàn)，這種賽事在未來的每年都會舉行。 對智能車輛的研究可以大大促進(jìn)其他學(xué)科的快速發(fā)展，如探索行星。美國火星車勇氣號和機(jī)遇號在探索火星和浩瀚的宇宙方面起到了不可替代的作用。[13,23].在中國，2000年11月中國政府發(fā)布了有關(guān)“中國航天”的白皮書，它的目標(biāo)是在不久的將來探索月球和其他行星。此外，空間移動機(jī)器人是探索行星的關(guān)鍵，這可能會受益于利用太陽能。 1.2智能汽車的關(guān)鍵技術(shù) 正如我們之前提到的，智能車是一套集環(huán)境感知和建模，定位和地圖構(gòu)建，路徑規(guī)劃和決策，以及運動控制于一身的多傳感器系統(tǒng)，如圖1.1所示。環(huán)境感知和建模模塊負(fù)責(zé)用于感測環(huán)境結(jié)構(gòu)的多感官方式，并提供周圍環(huán)境的模型。在這里，環(huán)境模型，包括一系列的運動對象，即靜態(tài)障礙物，車輛相對于當(dāng)前道路的位置，路的形狀，等等。最后，該模塊提供了環(huán)境模型和本地化地圖構(gòu)建模塊，這些模塊通過視覺，激光雷達(dá)和雷達(dá)處理原始數(shù)據(jù)。第二模塊，車輛定位和地圖創(chuàng)建模塊，是使用幾何特征來估計其在地圖中的位置以確定車輛的位置，并解釋傳感器信息來估算其在世界地圖中的位置。其結(jié)果是，第二模塊根據(jù)環(huán)境模型和局部映射得到全局地圖。路徑規(guī)劃和決策模塊的作用是確保車輛按照地面，安全性，舒適性，車輛動力學(xué)，和環(huán)境背景的規(guī)則操作。因此，該模塊可以提高任務(wù)效率和生成所需的路徑。最后的模塊，運動控制模塊是執(zhí)行車輛及其周圍環(huán)境之間相互作用的命令，以實現(xiàn)計劃的路徑。這些模塊的簡單介紹如下。 1.2.1基于多傳感器融合的環(huán)境感知與建模 圖1.2示出了一般的環(huán)境感知和建?？蚣?。從這個框架，我們可以看出：（i）原始數(shù)據(jù)是由不同的傳感器收集；（ii）從原始數(shù)據(jù)中可以提取不同的特征，如道路（對象）的顏色，車道邊，建筑輪廓；（iii）通過分類器認(rèn)知語義對象包括車道標(biāo)志，車輛，行人；（iv）我們可以在駕駛環(huán)境下推斷出車輛的位置。 1.多傳感器融合 多傳感器融合是智能車輛的基本框架，以便更好地感知周圍的環(huán)境結(jié)構(gòu)，以及探測物體/障礙的基本框架。粗略地說，用于周圍環(huán)境感知的傳感器分為2類：主動和被動的。有源傳感器包括激光雷達(dá)，雷達(dá)，超聲波和無線電，而常用的無源傳感器是紅外線和視覺照相機(jī)。不同的傳感器能夠提供不同的檢測精度和范圍，并對環(huán)境產(chǎn)生不同的影響。也就是說，結(jié)合各種傳感器不僅可以覆蓋短距離，而且長距離的對象/障礙也可以覆蓋，并在各種天氣條件下工作。此外，不同傳感器的原始數(shù)據(jù)可以被低級別，高級別和混合級別的傳感器融合。[4，14，20，35]. 2.動態(tài)環(huán)境建模 基于動態(tài)環(huán)境建模的車載攝像頭在智能車輛中起著重要的作用。[17]. 然而，由于自運動，模糊，光變化的綜合影響這是極具挑戰(zhàn)性的。因此，對于傳統(tǒng)的漸進(jìn)光照變化的方法，小型運動物體，如背景減除，工作的不是很好，即使是那些已被廣泛用于在監(jiān)視中的應(yīng)用程序。因此，越來越多的方法用來處理這些問題[2，17]. 不幸的是，以可靠的模型和更新背景,它仍然是一個懸而未決的問題。 當(dāng)在導(dǎo)航道路，十字路口，停車場，擁擠的十字路口要選擇不同的駕駛策略時，一些廣泛的場景通常被認(rèn)為在路徑規(guī)劃和決策。因此，場景估計有助于進(jìn)一步的決策，這經(jīng)常用于城市挑戰(zhàn)中。 3、目標(biāo)檢測與跟蹤 在一般情況下，在駕駛環(huán)境中，我們要注意的的是靜態(tài)/動態(tài)障礙，車道標(biāo)記，交通標(biāo)志，車輛和行人。相應(yīng)地，目標(biāo)檢測和跟蹤是環(huán)境感知與建模的關(guān)鍵部分。 1.2.2車輛定位與地圖創(chuàng)建 車輛定位與地圖構(gòu)建的目標(biāo)是通過結(jié)合環(huán)境模型、局部地圖和全局信息來生成全局地圖。在自動駕駛過程中，車輛定位是可以估計道路幾何形狀或相對于道路的車輛的已知地圖或未知的地圖。因此,車輛定位是指道路形狀估計,過濾,改變車輛構(gòu)成坐標(biāo)系。對于車輛定位，我們面臨的挑戰(zhàn)如下：（i）通常，來自GPS / DGPS的絕對位置會由于信號傳輸導(dǎo)致位置發(fā)生變化；（ii）路徑規(guī)劃和決策模塊需要的不僅僅是輸入車輛的絕對位置；（iii）傳感器噪聲大大影響車輛定位的準(zhǔn)確性。關(guān)于第一個問題，雖然在GPS和其變體已在車輛的定位得到廣泛應(yīng)用，它的性能可能會由于建筑物和樹木對信號的堵塞和反射而降低。在最壞的情況下,慣性導(dǎo)航系統(tǒng)(INS)可以保持一個位置的解決方案。至于第二個問題，本地地圖會融合激光，雷達(dá)和視覺數(shù)據(jù)與車輛狀態(tài)來用于定位和追蹤靜態(tài)/動態(tài)障礙物和車道。此外，全球地圖可能包含車道的幾何信息，車道標(biāo)線，標(biāo)牌步，停車場，檢查點和提供全球環(huán)境信息。談到第三個問題，各種噪聲模塊被視為降低定位誤差。[26].