GPS动态定位后处理软件研究
本文选题:动态定位 + 后处理 ; 参考:《西南交通大学》2013年硕士论文
【摘要】:实时动态定位(RTK)是GPS重要的应用研究领域之一,通过建立参考站与流动站间的无线通信链路获取运动目标的实时位置。但对于没有实时性要求的应用,可将采集到的数据带回室内进行后处理,即动态后处理。同时,精密星历的使用及更完善的误差模型修正也使得动态定位后处理在定位精度方面更具优势。Track是目前动态定位后处理的常用解算模块,它运行在UNIX或LINUX操作系统上,安装复杂且不具备可视化界面。 鉴于这些需求与不足,本文通过研读Track源代码,在掌握单历元动态定位理论的基础上针对Windwos操作系统开发了一套界面友好的动态定位后处理原型程序,并使用实测数据对其进行试验,分析了原型程序的模糊度解算能力和定位精度,讨论了不同基线长度与定位精度的关系、不同观测时长与定位精度的关系。 本程序的核心是整周模糊度解算和单历元卡尔曼滤波定位,包括以下功能模块:文件读取模块;参数设置模块;模糊度解算模块;卡尔曼滤波模块;其他辅助功能模块(文档输出、坐标系转换、时间系统转换、时序图显示、数据库查询等)。 本文通过八组实测IGS或四川CORS网观测1HZ数据展开试验。前四组数据具有相同观测时长(1小时),基线长度分别为5km、42kmm、123kmm、235km,用于分析程序的整周模糊度解算能力和不同基线长度的定位精度;后四组数据的测站相同、起始时间相同,观测时长分别为15分钟、30分钟、45分钟、60分钟,用于分析观测时长与定位精度的关系。试验方法为提取Track模块和本程序解算各组数据的整周模糊度和坐标结果,比较两程序解算的坐标均值和标准差,并将均值与测站理论坐标作比较。试验结论如下: 本程序的整周模糊度解与Track计算结果基本吻合,部分模糊度存在1-2周差异。Track处理1HZ数据,短基线的定位精度可达毫米级;中短基线水平精度可达毫米,高程精度为1-2厘米;中长基线的水平定位精度在1-2厘米,高程精度在2-3厘米;长基线的定位精度可达厘米级。本文开发的原型程序基本实现了动态定位后处理功能,短基线的定位精度可达1-2厘米;中短基线定位精度为2-3厘米;中长基线定位精度为厘米级别;长基线的定位精度在分米级。平面精度优于高程精度,基线越短解算精度越高。观测时间过短时后处理定位精度难以保证,30分钟以上初始观测时长且新加入卫星观测15分钟以上时,后处理可达厘米级定位精度;观测时间满足需要后,延长观测时间并不能提高定位精度。
[Abstract]:Real-time dynamic location (RTK) is one of the most important applications of GPS. The real-time position of moving target is obtained by establishing a wireless communication link between reference station and mobile station. However, for applications without real-time requirements, the collected data can be brought back to the room for post-processing, that is, dynamic post-processing. At the same time, the use of precision ephemeris and more perfect error model correction also make dynamic positioning post processing have more advantages in positioning accuracy. Track is a common solution module for dynamic positioning post processing, which runs on UNIX or Linux operating system. Complex installation and no visual interface. In view of these requirements and shortcomings, this paper, by studying track source code and mastering the theory of single epoch dynamic location, develops a set of friendly interface dynamic positioning post-processing prototype program for Windwos operating system. The ambiguity resolution ability and positioning accuracy of the prototype program are analyzed, and the relationship between different baseline lengths and positioning accuracy, and the relationship between different observation duration and positioning accuracy is discussed. The core of this program is integer ambiguity resolution and single epoch Kalman filter positioning, including the following functional modules: file reading module; parameter setting module; ambiguity resolution module; Kalman filter module; Other auxiliary function modules (document output, coordinate system conversion, time system conversion, timing chart display, database query, etc.). In this paper, eight groups of IGS or Sichuan cors data are used to test the 1HZ data. The first four groups of data have the same observation time (1 hour) and the baseline length is 5 km-1 / 42 kmm-1 123kmm-1 / 235km respectively, which is used to analyze the whole cycle ambiguity resolution ability of the program and the positioning accuracy of different baseline lengths. The last four groups of data have the same stations and the same starting time. The observation time is 15 minutes, 30 minutes, 45 minutes and 60 minutes, respectively, which is used to analyze the relationship between observation time and positioning accuracy. The test method is to extract the whole cycle ambiguity and coordinate result of track module and this program, compare the coordinate mean and standard deviation of the two programs, and compare the mean value with the theoretical coordinate of the station. The experimental results are as follows: the whole cycle ambiguity solution of this program is in good agreement with the track calculation results. The partial ambiguity has 1-2 weeks difference .Track deals with 1HZ data, and the positioning accuracy of the short baseline can reach millimeter level. The horizontal accuracy of the medium and short baselines is up to mm, the elevation accuracy is 1-2 cm; the horizontal positioning accuracy of the middle and long baselines is 1-2 cm and the elevation accuracy is 2-3 cm; the positioning accuracy of the long baselines can reach cm level. The prototype program developed in this paper basically realizes the post-processing function of dynamic positioning. The positioning accuracy of the short baseline can reach 1-2 cm, the positioning accuracy of the middle and short baseline is 2-3 cm, the positioning accuracy of the middle and long baseline is centimeter level. The positioning accuracy of the long baseline is at the decimeter level. The plane accuracy is better than the elevation precision, and the shorter the baseline is, the higher the accuracy is. If the observation time is too short and the post-processing precision is difficult to ensure the initial observation time is longer than 30 minutes and the satellite observation time is more than 15 minutes, the post-processing can reach the centimeter level positioning accuracy, after the observation time meets the need, Prolonging the observation time can not improve the positioning accuracy.
【学位授予单位】:西南交通大学
【学位级别】:硕士
【学位授予年份】:2013
【分类号】:P228.4
【参考文献】
相关期刊论文 前10条
1 祝会忠;高星伟;秘金钟;李明;;一种GPS整周模糊度单历元解算方法[J];测绘科学;2011年04期
2 喻国荣;单历元模糊度解算问题[J];测绘通报;2003年11期
3 柯福阳;王庆;余学祥;刘辉;;单历元GPS变形监测快速解算关键问题的研究[J];测绘通报;2008年03期
4 胡国荣,欧吉坤;改进的高动态GPS定位自适应卡尔曼滤波方法[J];测绘学报;1999年04期
5 董绪荣,陶大欣;一个快速Kalman滤波方法及其在GPS动态数据处理中的应用[J];测绘学报;1997年03期
6 周星;;基于GAMIT TRACK和Bernese PPP的地震监测比较[J];测绘信息与工程;2010年02期
7 阳仁贵,欧吉坤,任超;GPS单历元载波相位整周模糊度的解算方法研究[J];长江科学院院报;2005年01期
8 刘婵媛;陈国光;;基于GPS的卡尔曼滤波技术研究[J];弹箭与制导学报;2006年04期
9 谢世杰;GPS后处理动态法[J];地矿测绘;2004年04期
10 苏小宁;孟国杰;胡丛玮;伍吉仓;;基于TRACK进行GPS单历元定位[J];大地测量与地球动力学;2009年03期
相关硕士学位论文 前3条
1 苏小宁;高频GPS单历元定位方法及其在地震学中的应用研究[D];中国地震局地震预测研究所;2011年
2 雷礼平;GPS动态卡尔曼滤波算法研究[D];电子科技大学;2003年
3 袁志明;GPS单历元动态定位的研究[D];山东科技大学;2006年
,本文编号:2064023
本文链接:https://www.wllwen.com/kejilunwen/dizhicehuilunwen/2064023.html