带有区间不确定性的约束平差算法及应用

An algorithm with interval uncertain constraint in adjustment model and its application

针对测量数据中带有区间约束先验信息和附加信息,基于Kuhn-Tucker条件,将测量平差问题转化为二次规划问题,该文提出了一种处理参数带有区间不确定性的新算法,给出了算法具体模型和解算步骤,并且通过模拟数值实验和病态测边网数据计算,分析了在处理病态问题时,最小二乘平差的局限性,通过与岭估计和奇异值分解法的结果相比较,说明了参数带有区间不确定性的平差算法的有效性。

Considering the measurement data contains prior information and additional information.Based on Kuhn-Tucker condition used in quadratic programming,this paper propose a new algorithm with interval uncertain,and give its concrete model and solving steps.According to the results of simulated experiment and ill-posed trilateration net data,it shows that the least-squares has limitation when processing morbid problem.Compared with the results of ridge estimation and singular value decomposition（SVD）,it shows that the algorithm with interval uncertain of parameter is effective.