【摘要】：三維海底地形輔助導(dǎo)航技術(shù),具有操作直接,可視性好,交互性強(qiáng),信息量大等優(yōu)勢,使其廣泛應(yīng)用在導(dǎo)航領(lǐng)域中,尤其在變幻莫測的海洋復(fù)雜環(huán)境導(dǎo)航中,具有極其重要的地位,而建立大范圍的真實(shí)三維地形模型則成為實(shí)現(xiàn)這項(xiàng)技術(shù)不可或缺的重要內(nèi)容。本文主要研究基于真實(shí)的二維海圖提取的水深數(shù)據(jù)基礎(chǔ)上,實(shí)現(xiàn)真實(shí)海洋地形模型的重建,從而在視景仿真軟件MultiGen Creator中得到較為真實(shí)的地形,并最終在Vega Prime中實(shí)現(xiàn)虛擬海洋環(huán)境視景單元的仿真,及在該環(huán)境下的航行器漫游。主要的研究內(nèi)容如下:首先,本文介紹并分析海深數(shù)據(jù)的提取和處理,并從二維海圖中提取水深數(shù)據(jù),針對水深數(shù)據(jù)本身的缺陷,研究了基于IFS的插值算法,對其二維和三維進(jìn)行分析,并與其他算法進(jìn)行比較,最終得到需要的DEM模型。其次,本文設(shè)計(jì)由海深處理數(shù)據(jù)得到海底地形文件的技術(shù)路線。利用插值算法得到曲面文件,將此文件格式轉(zhuǎn)換成了Creator能使用的.ded格式,然后,采用三角網(wǎng)格法,在Creator中獲取三維地形文件,最終,通過紋理及渲染技術(shù)得到較為真實(shí)的海洋地形,同時(shí)利用Creator畫出了航行器模型。然后,針對海底地形路徑規(guī)劃問題,本文研究遺傳算法在路徑規(guī)劃上的應(yīng)用,著重分析評價(jià)函數(shù)、插入算子、懲罰因子等對路徑規(guī)劃的不同影響,并設(shè)計(jì)了適用于不同地形的路徑優(yōu)化算法。實(shí)驗(yàn)結(jié)果表明,路徑優(yōu)化算法中所提出的懲罰函數(shù)及啟發(fā)性知識的使用,能夠使算法根據(jù)不同地形的特點(diǎn)靈活的得到最優(yōu)路徑。最后,本文分析了仿真程序的設(shè)計(jì)過程和實(shí)現(xiàn)技術(shù),提出對仿真程序的漫游方式進(jìn)行改進(jìn),驗(yàn)證了建立的三維地形的可用性和地形數(shù)據(jù)庫的合理性,并利用Vega Prime2.01和VS2003實(shí)現(xiàn)三維海洋環(huán)境的可視化及航行器在海底避障漫游功能。
[Abstract]:3D seafloor terrain aided navigation technology has the advantages of direct operation, good visibility, strong interaction and large amount of information, which makes it widely used in the field of navigation, especially in the unpredictable marine complex environment navigation. It has an extremely important position, and the establishment of a large range of real three-dimensional terrain model has become an indispensable part of this technology. In this paper, based on the water depth data extracted from the real two-dimensional chart, the reconstruction of the real ocean terrain model is realized, so that the more real terrain can be obtained in the visual simulation software MultiGen Creator. Finally, the simulation of virtual marine environment scene unit is realized in Vega Prime, and the navigator roaming in this environment is realized. The main research contents are as follows: firstly, this paper introduces and analyzes the extraction and processing of sea depth data, and extract water depth data from two-dimensional chart. Aiming at the defects of water depth data itself, the interpolation algorithm based on IFS is studied. The two-dimensional and three-dimensional analysis and comparison with other algorithms are carried out, and the required DEM model is finally obtained. Secondly, this paper designs the technical route to obtain the seafloor terrain file from the sea depth processing data. The surface file is obtained by interpolation algorithm, and the file format is converted into .ded format that can be used by Creator. Then, the three-dimensional terrain file is obtained in Creator by using triangular grid method. Finally, The real ocean terrain is obtained by texture and rendering technology, and the navigator model is drawn by Creator. Then, aiming at the problem of seafloor terrain path planning, this paper studies the application of genetic algorithm in path planning, focusing on the different effects of evaluation function, insertion operator and penalty factor on path planning. A path optimization algorithm for different terrain is designed. The experimental results show that the penalty function and enlightening knowledge proposed in the path optimization algorithm can make the algorithm obtain the optimal path flexibly according to the characteristics of different terrain. Finally, this paper analyzes the design process and implementation technology of the simulation program, improves the roaming mode of the simulation program, and verifies the availability of the established 3D terrain and the rationality of the terrain database. Vega Prime2.01 and VS2003 are used to realize the visualization of 3D marine environment and the navigation function of navigator to avoid obstacles in the seafloor.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：P716

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 丁彪;;基于改進(jìn)遺傳算法的移動(dòng)機(jī)器人路徑規(guī)劃[J];自動(dòng)化應(yīng)用;2014年01期

，

本文編號：2487087