部分待估参数具有先验随机性的WTLS平差  

Weighted Total Least-Squares Adjustment with Partial Prior Random Parameter

作　　者：邓兴升 黄小鹏 彭思淳 DENG Xingsheng;HUANG Xiaopeng;PENG Sichun(School of Traffic and Transportation Engineering,Changsha University of Science and Technology, 960 Second South-Wanjiali Road,Changsha 410114,China)

机构地区：[1]长沙理工大学交通运输工程学院

出　　处：《大地测量与地球动力学》2018年第9期968-973,990,共7页Journal of Geodesy and Geodynamics

基　　金：公路地质灾变预警空间信息技术湖南省工程实验室基金(kfj150602);湖南省研究生科研创新基金(16101030004)~~

摘　　要：部分待估参数具有先验随机信息,且误差方程系数矩阵含有观测误差,是一类新的平差问题。本文构造了部分待估参数含有先验随机信息的加权整体最小二乘平差函数模型,推导该模型参数估计与精度评定公式,给出计算步骤,适用于一般情形。实例对比分析证明,该算法正确可靠,迭代收敛速度较优。A total least square with partial random parameter adjustment problem occurs if some of the estimated parameters in an adjustment problem have priori random information and the error equation coefficient matrix contains observation errors.This paper proposes a function model of total least squares adjustment with additional partial random parameters.The model has general adaptability.The algorithms formula of parameter estimation and accuracy evaluation are derived,and the steps of computation are presented,which can process the data that only partial(from 0 to all)parameter has random prior information.The feasibility,reliability and correctness of the algorithms are demonstrated by several examples and comparative analysis.The proposed algorithms have advantages in iterative convergence times.

关 键 词：参数估计 先验信息 加权整体最小二乘 随机参数 

分 类 号：P207[天文地球—测绘科学与技术]