基于遗传算法的广义非线性最小二乘测量平差  

Surveying and mapping adjustment of generalized least squares based on genetic algorithm

作　　者：宁伟[1] 张永常[1] 陈志荣[1] 欧吉坤[2] 

机构地区：[1]徐州师范大学计算机科学与技术学院,江苏徐州221116 [2]中国科学院测量与地球物理研究所,湖北武汉430077

出　　处：《辽宁工程技术大学学报（自然科学版）》2006年第B06期87-89,共3页Journal of Liaoning Technical University (Natural Science)

基　　金：国家自然科学基金资助项目(401740034020400140074003);江苏省教育厅高校自然科学研究计划基金资助项目(05KJD520226);徐州师范大学自然科学研究基金资助项目

摘　　要：为了解决当前“数字地球”、“数字国家”、“数字城市”、“数字矿山”等数字化工程中广泛出现的具有多源、多维、多类型、多精度、动态和非线性等特点的测量数据的误差处理问题,本文采用了一类基于自然界生物进化基本法则而发展起来的新算法——遗传算法,在提出、设计基于遗传算法的广义非线性最小二乘参数平差方法的同时,给出了遗传算子中选择、交叉、变异算子的设计,以及具体的算法步骤。通过实例计算表明,该遗传算法是进行广义非线性最小二乘参数估计的具有全局最优化的有效方法,为广义非线性测量数据处理提供了又一新的思路。In order to resolve the questions on surveying errors from multi-original, multi-dimensional,multi-kind, multi-precise, dynamic and nonlinear surveying data, which come from the digital engineering of digital earth, digital nation, digital city and digital mine, a new algorithm is used, which is based on biology evolution, i.e. genetic algorithm. Nonlinear least square based on genetic algorithm is put forward ,and design of the selection, the cross and the variance of the genetic arithmetic operators and the algorithm steps are introduced. The method is an effective and whole optimizing algorithm to do nonlinear least square estimated parameter from some examples, and a new idea is provided for nonlinear least square.

关 键 词：遗传算法 广义非线性最小二乘法 参数估计 

分 类 号：P207.2[天文地球—测绘科学与技术]