汽車導航系統(tǒng)畢業(yè)論文中英文資料外文翻譯文獻

1、汽車導航系統(tǒng)中英文資料外文翻譯文獻 使用gis數(shù)據(jù)庫和激光掃描技術(shù)為汽車導航系統(tǒng)獲取路標索引現(xiàn)在的汽車導航系統(tǒng)以地圖，圖形，以及聲音的形式提供給用戶行駛中的信息，然而他們還遠遠不能支持基于道路標記的導航，而這也是對我們來說更簡單的導航理念，并且這也在不久要實現(xiàn)的個人導航系統(tǒng)中占據(jù)重要的位置。為了提供這樣的一種導航，第一步就要識別恰當?shù)牡缆窐擞浾б豢此坪鹾芎唵?，但是如果考慮到要把覆蓋了歐洲、北美、日本大部分地區(qū)的信息傳輸給數(shù)據(jù)庫的挑戰(zhàn)，我們就有理由自命不凡了。在這里，我們將講解從已存在的gis數(shù)據(jù)庫中獲取道路標記的方法。因為這些數(shù)據(jù)庫大多數(shù)沒有包含建筑物的高度和視圖信息，我們將展示這些信息怎樣從

2、激光掃描數(shù)據(jù)中分離出來。1簡介1995年在上層階級的汽車里汽車導航系統(tǒng)就已經(jīng)出現(xiàn)了，而且現(xiàn)在幾乎可以在任何樣式的汽車中找到導航系統(tǒng)。他們是相對復雜和成熟的系統(tǒng)可以以數(shù)字地圖，行駛方向圖形，以及行駛中的聲音信息提供路線導航?；厮?980年汽車導航系統(tǒng)開始興起的時候，一些大的問題都得到了解決：例如絕對位置，適合導航的大量地圖的提供，快速算路以及可靠的路線導航。然而，傳送這些信息的原始概念并沒有得到較大的改善。聲音的導航仍然用相對小的提示：（例如 現(xiàn)在向右轉(zhuǎn)），這只涉及到了道路分布的屬性。這不是最理想的，因為1）路線分布的特征在較大距離的時候是不可見的，這是因為司機受局限的位置以及視角，2）人們最習

3、慣的導航方式是通過道路標記，也就是沿路中一系列的可識別可記憶的的圖像的提供。很明顯，作為道路標記的建筑物的提示與聲音提示結(jié)合起來，將是導航發(fā)展中更人性化的一個方向，就像我們下邊討論的那樣，這將很好的集成到今天的汽車導航系統(tǒng)中去因為不意味著對系統(tǒng)和數(shù)據(jù)結(jié)構(gòu)的大的改動。所以，主要的問題在于識別合適的道路標記以及估計他們對于導航提示的可用性。這里，我們將解釋已存的數(shù)據(jù)庫怎樣開發(fā)以解決第一個問題，而激光瀏覽數(shù)據(jù)庫將解決后一個。2使用激光掃描數(shù)據(jù)集的可見性分析2.1可見性分析如果我們以來自激光掃描的dsm上直觀的可見性作為分析的基礎(chǔ)，我們會做的更好。我們將不會獲得像當初估計的那樣使建筑物從任何的地點都被

4、清晰地看見。 我們依照下列各項的方式，對于任何的觀察點的位置和觀看方向定義給予的水平線和垂直的一個虛擬的照相機的外部方位看角。 這個虛擬的照相機表示為駕駛者的視野。高度起源于dsm本身,然而看角從 gdf 數(shù)據(jù)組中對應的街道的固定方位被獲得。虛擬圖像的平面然后被光柵過濾, 每個圖素定義物體空間的一道光線。所有的光線在物體空間中被追蹤并且用 dsm 決定交集。對于每次擊中，對應的物體數(shù)據(jù)被試映圖的虛擬圖像查詢獲得。雖然這個方法與 光線追蹤 類似并用在計算機圖形方面和平時假設(shè)計算中，但是自從我們只在光線的第一個擊中方面感興趣之后，它實際上相當速，而且dsm只是2.5 d,虛擬圖像的飛機以此下去可能

5、從底部到頂端被有效率地計算,向逐漸增加的物體空間進軍。2.2 追蹤可見性在最后的一個區(qū)段中，單一視野被計算。 然而,道路標記被一個路線排定指令選擇，而且一定在整個調(diào)遣期間是看得見的。這可能沿著對應的調(diào)遣定義的軌道追蹤物體的可見性。對于我們的第一次實驗，我們使用只有一個自然的附近區(qū)域作為可見, 即被虛擬圖像上的飛機的對應物體覆蓋的區(qū)域。 圖1顯示了一個例子。 我們假設(shè)白色的多角形是我們駕駛者使用的軌道。然后問題是如果以其他的方法識別它是一個道路標記的城鎮(zhèn)大廳,是一個可以被道路標記唯一表示的適當物體。對這一次的結(jié)論，我們的運算法則是追蹤整個的軌道,以等距離隔開的位置和在軌道旁邊的固定方位產(chǎn)生罰款者

6、眼中虛擬的視野。對于每個如此的視野，虛擬圖像中的飛機上的每個物體覆蓋的區(qū)域被決定。圖 2顯示沿著圖1的軌道所有的那些區(qū)域的一種情況。當物體出現(xiàn)的時候, 能產(chǎn)生典型的有遮掩的曲線, 變化比較大和最后消失的就如被途徑人所觀察。 在這種特別的情形中，當位置是城鎮(zhèn)門廳之前并留下狹窄的街道和進入廣場的時候，或視野弄寬的時候，許多物體在附近看變成規(guī)格為65號。為了確定城鎮(zhèn)大廳是否為一個適當?shù)奈矬w, 從圖2上的對應的曲線看，從規(guī)格65到115號是最大的,也就是城鎮(zhèn)大廳是駕駛者視野中最大的物體。而且，曲線比從規(guī)格13號開始的更大,這意味城鎮(zhèn)大廳是一少部分大約在進入廣場的之前100 公尺處被看見( 可能是決定點

7、) 因此，在這種情況下我們既能查出顯示比較大的物體，也能在駕駛者的視野中將他最早顯示出來。圖1 俯視圖上的軌跡線圖2：基于框架數(shù)字的可見性劃分3數(shù)據(jù)地圖汽車導航系統(tǒng)使用的地圖不僅包含幾何學和道路網(wǎng)絡的連接性而且包含了大量的關(guān)于物體，屬性和關(guān)系的附加信息。一個好的觀點能夠從歐洲的標準獲得，舉例來說，（年月的地理數(shù)據(jù)文件），其中包括了博物館，戲院，文化中心和市政廳等的信息。地圖數(shù)據(jù)是被諸如電子地圖的地圖數(shù)據(jù)庫廠商獲得并通過交換的方式提供給汽車導航系統(tǒng)生產(chǎn)商的（例如）。在那里，它被轉(zhuǎn)換到最后在地圖激光唱碟或數(shù)字化視頻光上被發(fā)現(xiàn)的專有格式。數(shù)據(jù)必須從一種描述形式轉(zhuǎn)換成被汽車導航系統(tǒng)支援的另一種被特殊化

8、的形式，這轉(zhuǎn)變是高度非凡的。時常，結(jié)構(gòu)和價值被這個轉(zhuǎn)換過程預先計算了，目的是為了要減輕航行系統(tǒng)的在線資源 , 例如帶寬和時間。這個模塊的其中一部分也是為每個十字路口產(chǎn)生一個點陣式，目的是描述所有的可能轉(zhuǎn)向的組合。在汽車導航系統(tǒng)中使用了眾所周知的箭頭符號來標識，這就需要所有道路的十字路口的交匯情況將被存儲。在轉(zhuǎn)向過程中，對于帶有路標的汽車導航系統(tǒng)的附加信息會被完整化。在本文中，概括說明了是怎樣通過與地圖數(shù)據(jù)和激光掃描數(shù)據(jù)結(jié)合來確定道路幾何圖形的適合的路標，重要的一點是那些附加的數(shù)據(jù)信息僅僅在這個轉(zhuǎn)換過程中被使用。在那之后，僅僅是基于路標的行使指示還存在，這些是行使指示可能在一種非常緊湊的形式下被

9、編碼，并且要與每一個十字路口各自的已被存儲在專有地圖格式的數(shù)據(jù)信息相協(xié)調(diào)。因此，路標技術(shù)的整合沒有在現(xiàn)在的汽車導航系統(tǒng)中造成障礙，這些主要問題是來自那些用自動或半自動方法的指令中的。4 激光掃描和城市模型在二十世紀九十年代，靠空氣傳播的激光掃描作為獲得表面的模型的新方法變得可用。隨后，掃描系統(tǒng)提高了并且指引全球范圍也因為足夠的精度變得可行。今天，靠空氣傳播的激光掃描是一項成熟的技術(shù)為大多數(shù)公司提供系統(tǒng)和服務。掃描很大的區(qū)域是可能的，例如整個荷蘭已經(jīng)被掃描過了，德國的baden-wurttemberg州也正在進行掃描，他們中每一個的面積都超過了30平方千米。天線激光掃描機直接地生產(chǎn)地球的表面密集

10、的點云 (baltsavias et al。,1999). 他們對獲得密集的都市區(qū)域的數(shù)傳表面模型 (dsms) 是特別地適當?shù)? 如同他們保存跳躍邊緣一樣相當好。 大多數(shù)的系統(tǒng)能夠測量不只有高度, 也有反射系數(shù), 和首先,最后的或多樣的回行脈沖,他們允許分開樹形天篷和地面。 (kraus 和 rieger,1999)主要的問題是怎樣從激光掃描數(shù)據(jù)組中獲取關(guān)于人造結(jié)構(gòu)的符號信息，可能和天空的或陸地的圖像聯(lián)合。尤其, 自動機械世代的城市模型是而且仍然是一個強烈的研究領(lǐng)域, 這個討論是超過本文的范圍的。 在這一問題上，讀者可以咨詢“ascona 工作室”的優(yōu)秀的成果。 (grun et al.,

11、1995, grun et al., 1997,baltsavias et al., 2001).然而，實質(zhì)性研究努力還是很必要的直到高度自動化的物體獲取系統(tǒng)可以可靠地工作。另一方面，三維空間存在的物體信息在今天存在的gis數(shù)據(jù)庫中還遠遠不是普遍的。所以，在本文中我們將考慮把gis數(shù)據(jù)庫和激光掃描dsms聯(lián)合起來在一個圖標層上，不明確地重建物體的三維空間的形狀而當做分開實體。圖3展示了一個數(shù)據(jù)資源被用過的例子，來自正在激光掃描的dsm，使有規(guī)則到1米的格子，街道的幾何形狀用從一個gdf數(shù)據(jù)組合的中心線表示，而建筑物的輪廓用從地籍圖上獲得的中心線表示。圖3 激光掃描5 結(jié)論及前景在本文中，已經(jīng)概

12、略說明路標是如何被取得的并且評估使用已存在的 gis 和激光掃描數(shù)據(jù)。 至于路標的取得，我們已經(jīng)調(diào)查基于顯示突出建筑物的二種不同的方法。 為了評估導航引導的有用性,我們用了基于來自激光掃描的 dsm 數(shù)據(jù)的一項可見性分析。 數(shù)據(jù)挖掘程序必須用真正的數(shù)據(jù)組來測試。 如果他們在現(xiàn)實世界中使用適當?shù)穆窐艘龑?，這個結(jié)論將會被證實。 除此之外，分析程序必須被擴展到不同的事物類型 (交通建筑,公園,體育運動設(shè)備等.) 從 atkis 數(shù)據(jù)提取舉例來說明。不同種類事物的數(shù)據(jù)預處理方法和當不同數(shù)據(jù)挖掘運算法則被提供到相同數(shù)據(jù)時產(chǎn)生的問題必須被調(diào)查。萃取的路標的可靠性不得不通過質(zhì)量測試來決定，目的是為了避免不明

13、確的目標誤導用戶。更多的依靠路線來決定路標的問題必須被調(diào)查: 用戶行駛方向和路標質(zhì)量可見性的影響。當我們只用了 虛擬的圖像大小 來估價一個事物的可見性時,有很大的空間來進步。舉例來說，如果一個事物被它前面或附近的事物擋住了，或者是整個輪廓的一部分，從虛擬的圖像，就能獲得遠距離的信息。首先激光掃描測量的脈搏能夠被整合，目的是為了獲得一個比較好的由樹導致阻塞的近似值。dsm 也可能被用來提供萃取的附加信息，例如，小塔被它前面的大建筑物擋住這個信息將被確定。跟蹤可見性的執(zhí)行使用等距離的時間取樣來代替空間取樣，這是基于車輛在臨近交叉路口的速度的。最后，存在于gdf數(shù)據(jù)中的poi數(shù)據(jù)被使用到可見性分析的

14、擴展是非常有趣的。附件2：外文原文extracting landmarks for car navigation systems using existing gis databases and laser scanningabstracttodays car navigation systems provide driving instructions in the form of maps, pictograms, and spoken language. however, they are so far not able to support landmark-based navigat

15、ion, which is the most natural navigation concept for humans and which also plays an important role for upcoming personal navigation systems. in order to provide such a navigation, the first step is to identify appropriate landmarks a task that seems to be rather easy at first sight but turns out to

16、 be quite pretentious considering the challenge to deliver such information for databases covering huge areas of europe, northern america and japan. in this paper, we show approaches to extract landmarks from existing gis databases. since these databases in general do not contain information on buil

17、ding heights and visibility, we show how this can be derived from laser scanning data.1 introductionmodern car navigation systems have been introduced in 1995 in upper class cars and are now available for practically any model. they are relatively complex and mature systems able to provide route gui

18、dance in form of digital maps, driving direction pictograms,and spoken language driving instructions (zhao, 1997).looking back to the first beginnings in the early 1980s, many nontrivial problems have been solved such as absolute positioning, provision of huge navigable maps, fast routing and reliab

19、le route guidance.however, the original concept of delivering the instructions has not changed very much. still, spoken language instructions use a relatively small set of commands (like turn right now), which only refer to properties of the street network. this is not optimal, since i) features of

20、the street network typically are not visible from a greater distance due to the low driver position and small observing angle, and ii) the most natural form of navigation for humans is the navigation by landmarks, i.e. the provision of a number of recognizable and memorizable views along the route.

21、obviously, the introduction of buildings as landmarks together with corresponding spoken instructions (such as turn right after the tower) would be a step towards a more natural navigation. as we argue below, this would be well integrable into todays car navigation systems as it would not imply a ma

22、jor modification of systems and data structures. thus, the main problem lies in identifying suitable landmarks and evaluating their usefulness for navigation instructions. in this paper, we show how existing databases can be exploited to tackle the first problem, while laser scanning data can be use

23、d to approach the second.2visibility analysis using laser scanning datasets2.1 visibility analysiswe can do better if we base the visibility analysis directly on the dsm from laser scanning. we will not obtain “beautiful” visualizations but instead a rather good estimate on which buildings can be se

24、en from any viewpoint (fig. 4(c). we realized this approach as follows. for any viewpoint, the position and viewing direction define the exterior orientation of a virtual camera of given horizontal and vertical viewing angle. this virtual camera represents the drivers view. the height is derived fro

25、m the dsm itself, whereas the viewing angle can be obtained from the orientation of the corresponding street segment in the gdf dataset.the virtual image plane is then rastered, each pixel defining a ray in object space. all the rays are traced in object space to determine intersections with the dsm

26、. for each hit, the corresponding object number is obtained by a lookup in an image containing rastered ground plan ids. although this method is similar to “ray tracing” used in computer graphics and often assumed to be computationally expensive, it is actually quite fast since (a) we are interested

27、 only in the first hit of the ray, and (b) the dsm is 2.5d only, so each column in the virtual image plane can be computed efficiently from bottom to top, marching in increasing distance in object space.2.2 tracking visibilityin the last section, visibility was computed for a single view. however, l

28、andmarks selected for a routing instruction must be visible during the entire manoeuvre. this can be checked by tracking the visibility of objects along the trajectory defined by the corresponding manoeuvre. for our first experiment, we use only a crude approximation for the visibility, namely the a

29、rea covered by the projection of the corresponding object on the virtual image plane. figure 1 shows an example. we assume that the white polygon is the trajectory we want the driver to use. the question then is if the town hall, identified to be a landmark by the methods of section 5, is a suitable

30、 object which can be used in a landmark-based instruction such as pass to the right of the town hall. to this end, our algorithm traces the entire trajectory, generating virtual views at equidistantly spaced positions and in the orientation de-fined by the trajectory. for each such view, the area co

31、vered by each object on the virtual image plane is determined. figure 2 shows a plot of all those areas along the trajectory of figure 1. one can see the typical peaked curves generated as objects appear, grow larger and finally disappear as the viewing position passes by. in this special case, one

32、sees also that many objects become visible around frame number 65, which is when the view widens as the position leaves the narrow street and enters the plaza in front of the town hall. in order to answer if the town hall is a suitable object, a look on figure 2 reveals that the corresponding curve

33、(shown in bold red) is largest for frame numbers 65 to 115 (with a small exception around frame 100), i.e. the town hall is the largest object in the drivers view. moreover, the curve is larger than zero starting from frame number 13, which means that the town hall is at least partly visible about 1

34、00 meters ahead of the position where the plaza is entered (which could be a decision point). thus, in this case we can verify both that the object appears large and that it appears early enough in the drivers view.figure 1:example trajectory, top view.figure 2: visibility plotted over frame number3

35、 digital mapsthe maps used by car navigation systems not only contain the geometry and connectivity of the road network but also a huge amount of additional information on objects, attributes and relationships. a good overview can be obtained from the european standard gdf, see e.g. (geographic data

36、 files 3.0, 1995). of particular interest are points of interest (poi) which include museums, theaters, cultural centers, city halls, etc.map data is acquired by map database vendors such as tele atlas or navtech and supplied to car navigation manufacturers in an exchange format (such as gdf). there

37、, it is converted to the proprietary formats finally found on the map cd or dvd. this conversion is highly nontrivial since the data has to be transformed from a descriptive form into a specialized form supporting effi-cient queries by the car navigation system. often, structures and values are prec

38、omputed by this conversion process in order to relieve the navigation systems online resources such as bandwidth and cpu time.part of this process is also to generate a matrix for each intersection which describes all possible turn combinations. also, for the well-known arrow pictograms used by car

39、navigation systems, the angles between all streets joining at an intersection are stored. it is during this conversion process where additional information for landmark-based navigation can be integrated. in this paper, we outline how the street geometry given by gdf can be combined with information

40、 from a cadastral map and laser scan data to identify suitable landmarks. an important point is that the additional datasets are used only during the conversion process. after that, only landmark-based driving instructions remain, which can be coded in a very compact form and are compatible with the

41、 per-intersection information already stored in proprietary map formats. thus, the technical integration of landmark-based instructions into current car navigation systems poses no major obstacles, and the main problem is to derive those instructions in some automatic or at least semiautomatic way.4

42、 laser scanning and city modelsduring the 1990s, airborne laser scanning became available as a new method for obtaining surface models. subsequently, the scanning systems were improved and direct georeferencing became feasible with sufficient accuracy. today, airborne laser scanning is a mature tech

43、nology with a multitude of companies offering systems and services (baltsavias, 1999). scanning of very large areas is possible, for example the entire netherlands have been and germanys state of baden-wurttemberg is in the progress of being scanned, each with an area of over 30.000 km2. aerial lase

44、r scanners produce dense point clouds of the earths surface directly (baltsavias et al., 1999). they are particularly suitable for obtaining digital surface models (dsms) in dense urban areas, as they conserve jump edges quite well. most systems are capable of measuring not only the height, but also

45、 the re-flectance, as well as first, last or multiple return pulses, which allows to separate tree canopy and ground (kraus and rieger,1999).the main problem is how to extract symbolic information about man-made structures from laser scanner datasets, possibly combined with aerial or terrestrial ima

46、ges. especially, the automatic generation of city models has been and still is an intense research field, the discussion of which is beyond the scope of this paper. the reader is referred to the excellent proceedings of the “ascona workshops” on this topic (grun et al., 1995, grun et al., 1997,balts

47、avias et al., 2001).however, there is still substantial research effort necessary until highly automated object extraction systems working reliably become available. on the other hand, three-dimensional object information is still far from being common in todays existing gis databases. in consequenc

48、e, in this paper we consider using two-dimensional gis databases in combination with laser scanner dsms on an iconic level, without explicitly reconstructing the three-dimensional shape of the objects as separate entities. figure 3 shows an example of the data sources used, which is a dsm from laser

49、 scanning, regularized to a 1 m grid, the street geometry represented by center lines from a gdf data set, and the outline of buildings from a cadastral map.figure3:laser scan5 conclusion and outlookin this paper, we have outlined how landmarks can be extracted and evaluated using existing gis and l

50、aser scanning data. as for the extraction, we have investigated two different methods based on data mining to reveal prominent buildings. in order to evaluate the usefulness for navigation instructions, we used a visibility analysis based on dsm data from laser scanning. both data mining procedures have still to be tes