精密单点定位相关技术问题研究

发布时间：2018-08-02 14:49

【摘要】：GPS精密单点定位一直是一个热门话题,从它的出现到之后的发展,许多的学者都在进行研究,由于该技术的应用越加的广泛,该定位方法具有操作简便,不依赖其他参考站的特点,在近地卫星定轨,国土调查等领域都应很好的应用,因此,研究精密单点定位具有很好的理论意义和实用价值。精密单点定位要求得到精密的卫星轨道和卫星钟差信息通过IGS提供,进一步需要进行星历内插和钟差内插,以得到所需时刻的信息,目前已经可以采用多种方法进行内插,并能达到比较高的精度,在卫星距离求解中需要得到精确的卫星距离,这需要考虑测站相关误差、卫星相关误差影响以及传播路径相关误差,需要进行地球自转改正、对流层延迟改正、卫星天线相位偏差、电离层延迟改正等多种误差改正。另外,由于信号在传播过程中可能发生周跳,或者出现比较大的粗差,因此,必须对数据进行预处理,以获得更好的数据源。在精密单点定位中,还要涉及到参数估计,其目的是解算待求参数,目前有卡尔曼滤波,最小二乘、序贯平差等多种方法。在学习了精密单点定位的理论基础知识后,通过对整个流程的仔细分析,梳理,开始了对GPS精密单点定位的研究,本文的主要研究内容如下：第一：探讨了用于精密单点定位的三种定位模型：传统模型、UofC模型、无模糊度模型,并分析了进行精密单点定位需要进行的误差改正模型,研究了数据预处理的方法,对参数估计方法进行了介绍和比较。第二：运用高次差法,相位减伪距法、宽巷相位减窄巷伪距法,消电离层法四种方法进行周跳探测实验,比较分析了四种方法在数据预处理方面的精度和使用范围上的优缺点。并提出在本文使用宽巷相位减窄巷相位与电离层残差法联合进行数据与处理能较好的探测出周跳。第三：结合实验数据分析了地球自转模型改正、相对论效应,对流层延迟等的改正大小,并研究了影响误差大小的主要因素。第四：比较不同滤波初值对卡尔曼滤波的影响,根据实验数据,研究了噪声阵、过程噪声矩阵、增益矩阵、预测协方差阵之间的关系。第五：本文以matlab7.0作为编程工具分别使用最小二乘平差方法和卡尔曼滤波算法对数据进行处理,卡尔曼滤波解算精度高,并且在收敛后单历元解算精度高,因此不仅适合静态单点定位还适合动态单点定位。分别使用快速星历和最终星历进行卡尔曼滤波计算,对两种星历进行比较。
[Abstract]:GPS precise single point positioning is a hot topic all the time. From its appearance to its development, many scholars are studying it. Because of the wide application of this technology, the method is easy to operate. Independent of the characteristics of other reference stations, it should be well applied in the field of near-Earth satellite orbit determination, land survey and so on. Therefore, the study of precise single-point positioning has good theoretical significance and practical value. Precise single point positioning requires precise satellite orbit and satellite clock difference information to be provided through IGS. It is further necessary to carry out ephemeris and clock difference interpolation to get the required time information. At present, many methods can be used to interpolate the satellite orbit and satellite clock difference. The accuracy of satellite distance should be obtained in the solution of satellite distance. It is necessary to take into account the errors of station correlation, satellite correlation error and propagation path, and to correct the rotation of the earth. Tropospheric delay correction, satellite antenna phase deviation, ionospheric delay correction and so on. In addition, due to the possibility of cycle slip or large gross error in the process of signal propagation, the data must be preprocessed to obtain a better data source. In precise single point positioning, parameter estimation is also involved, and its purpose is to calculate the parameters to be solved. At present, there are many methods such as Kalman filter, least square, sequential adjustment and so on. After studying the basic theoretical knowledge of precise single point positioning, through the careful analysis of the whole process and combing, the research on GPS precision single point positioning is started. The main contents of this paper are as follows: first, three kinds of positioning models for precise single point positioning are discussed: the traditional model of UofC, the model of no ambiguity, The error correction model of precision single point positioning is analyzed, the method of data preprocessing is studied, and the method of parameter estimation is introduced and compared. Second, four methods are used to detect cycle slips, such as high-order difference method, phase de-pseudo-range method, wide-lane phase de-pseudo-range method, and ionospheric elimination method. The advantages and disadvantages of the four methods in data preprocessing accuracy and application range are compared and analyzed. It is also proposed that the data processing and data processing by using the wide lane phase to reduce the narrow lane phase and the ionospheric residuals can detect the cycle slip well in this paper. Third, the correction magnitude of the Earth rotation model correction, relativistic effect and tropospheric delay are analyzed based on the experimental data, and the main factors affecting the error are studied. Fourthly, the effects of different initial filtering values on Kalman filtering are compared. Based on the experimental data, the relationships among noise matrix, process noise matrix, gain matrix and prediction covariance matrix are studied. Fifthly, this paper uses matlab7.0 as programming tool to process the data using least square adjustment method and Kalman filter algorithm, which has high accuracy and high precision of single epoch solution after convergence. Therefore, it is not only suitable for static single point positioning, but also suitable for dynamic single point positioning. The fast ephemeris and the final ephemeris are used to calculate the Kalman filter, and the two ephemeris are compared.
【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】