部分最小二乘平差的快速多维粗差定位定值法  

Method of Partly Least Squares in Fast Gross Error Location and Estimation

作　　者：林国庆[1] 范东明[1] 谢用[1] 杨智博[2] 

机构地区：[1]西南交通大学测量工程系,四川成都610031 [2]新疆生产建设兵团勘测规划设计研究院,新疆乌鲁木齐830000

出　　处：《测绘》2010年第4期162-164,共3页Surveying and Mapping

摘　　要：根据部分最小二乘平差原理把观测值按照是否含有粗差分成两组,对含有粗差和不含粗差组分别实施最小二乘平差,从而实现粗差的定位定值。针对原部分最小二乘平差法在分组时速度慢、效率低的缺点,利用观测值改正数的特点,改进了分组方法,从而加快了分组速度,提高了粗差探测效率。According to the principle of partly least squares, the observed values according to whether they contain gross errors were divided into two groups. Containing gross errors and non groups were implemented least-squares adjustment, in order to achieve gross error location and estimation~ For the shortcomings of original partly least squares, this paper use the features of observations＇ corrections to improve grouping method and gross error detection efficiency.

关 键 词：部分最小二乘 观测值改正数 可靠性理论 粗差 定位 定值 

分 类 号：P207[天文地球—测绘科学与技术]