遗传算法在非线性最小二乘平差中的应用  

APPLICATION OF GENETIC ALGORITHMS IN NONLINEAR LEAST SQUARES ESTIMATION

作　　者：王穗辉[1,2] 

机构地区：[1]同济大学测量与国土信息工程系 [2]现代工程测量国家测绘局重点实验室,上海200092

出　　处：《大地测量与地球动力学》2006年第2期95-98,共4页Journal of Geodesy and Geodynamics

基　　金：徕佧测绘基金项目

摘　　要：为克服线性化经典平差的不足,尝试利用遗传算法全局和局部搜索力强的优势,进行非线性最小二乘参数估计。对遗传算法涉及的六要素及其非线性估计的精度评定等作了研究和分析。最后通过算例验证了其处理非线性问题的有效性。On the basis of the superiority of genetic algorithms whose ability of searching for the whole or the part is strong, nonlinear least squares estimation was used to overcome the disadvantage of classical least squares. 6 essential factors related to genetic algorithms and its precision evaluation of nonlinear estimation have been studied. The calculating results show that it is reasonable using the genetic algorithms to process the nonlinear estimation.

关 键 词：非线性最小二乘估计 遗传算法 六要素 适应度 精度评定 

分 类 号：P207.1[天文地球—测绘科学与技术]