节点文献

水下地形辅助导航方法研究

The Research of Terrain-aided Underwater Navigation

【作者】 辛廷慧

【导师】 徐德民;

【作者基本信息】 西北工业大学 , 控制理论与控制工程, 2004, 硕士

【摘要】 由于水下环境的非结构性、复杂性、不稳定性等特点,使得水下导航比较困难。惯性导航系统是实现水下导航的一种重要方法,但惯性导航存在累计误差,为了修正惯性导航的累积误差,提高导航精度,本文将飞机、导弹等所使用的地形辅助导航系统,应用到水下航行器上,成功地满足了自主水下航行器的导航需求。主要研究内容包括: 一、总结分析了当前水下导航的常用方法和不足之处,指出基于航位推算的导航方法,存在着最大的问题就是导航误差的积累问题,如果没有修正系统,这种方法很难满足使用要求;其次指出基于声学的导航方法,造价昂贵,导航范围有限,难以满足自主水下航行器大范围导航的需求; 二、总结了当前水下地形辅助导航的主要技术路线,一是基于地形高程的匹配方法,这是当前研究的重点;二是基于地形图像的匹配方法,由于缺乏良好的图像传感器,和可用的海底图像地图,目前这种方法还有待研究发展; 三、将两种地形高程匹配的常用方法应用到自主水下航行器上:以相关算法为前提的地形轮廓匹配(TERCOM)方法和以卡尔曼滤波为基础的桑迪亚惯性地形辅助导航(SITAN)方法,前者算法简单可靠,但是导航精度不高,后者虽然精度比较高,但存在着算法较为复杂、有可能发散和出现奇异值等问题。 四、在以上研究的基础上提出了几种改进的新方法,以提高导航的精度和可靠性。①采用并行SITAN方法来提高导航精度,并克服奇异值问题;②提出了TERCOM+SITAN方法,综合利用两者的优点,在保持SITAN导航精度的前提下,有效地克服了SITAN的两个缺点;③提出了一种基于连续蒙特卡洛滤波(常被称为Particle Filter)的地形匹配算法(PFTAN),有效地克服了利用SITAN时由于地形随机线性化带来的误差,使导航精度有较大的提高;④提出了TERCOM作为监视器的地形辅助导航思想,并将其应用到连续蒙特卡洛方法上,较大地增加了系统的可靠性和精度。

【Abstract】 As the cumulate error, inertial navigation system can’t provide perfect position and navigation for the Autonomous Underwater Vehicles (AUV) in unstructured, complex and instable ocean in a long time. The application of terrain-aided navigation techniques to the updating of an inertia navigation system has been investigated. In this paper, the main research works are as follows:1) Survey and analysis previous work in AUV navigation and summarize: As the cumulate error, the dead-reckoning and inertial navigation systems can’t be used without other modified system; the acoustic navigation system is used only in small range, and is very expensive.2) Two important approaches in geophysical navigation techniques are summarized, one is terrain contour based navigation, this approach is investigated keystone today; the other is image based navigation, due to no perfect image sensors and image seabed map, this approach has seldom been practised.3) Two important terrain contour based navigation algorithms has been applied to AUV, one is match algorithm-based Terrain Contour Match (TERCOM); the other is Kalman filter-based Sandia Inertia Terrain-aided Navigation (SITAN). The first algorithm is low precise but simple and credible, the second is high precise but complex and incredible.4) Developed four kinds of methods aimed to improve precision and credibility of navigation system. The first is Parallel Sandia Inertia Terrain-aided Navigation (PSITAN); The second is TERCOM+SITAN, it can restrain two important disadvantages of SITAN; The third is Particle filter-based Terrain-aided Navigation (PFTAN), the particle filter can reduce the error of navigation; The last is TERCOM+PFTAN, where TERCOM is looked as monitor to ensure the credibility of navigation system.

  • 【分类号】U675
  • 【被引频次】46
  • 【下载频次】1235
节点文献中: