参数带有区间约束的平差模型迭代算法

测量平差模型中的参数通常存在一些不确定的附加信息或先验信息，充分利用它们可以对部分参数进行约束，从而保证参数解的唯一性和稳定性。本文主要研究参数带有区间约束的平差模型。即，利用矩阵正则分裂方法，将平差问题转化成一个简单的二次规划问题，建立了一种新的参数估计迭代算法，并证明了算法的收敛性。最后通过实例说明了新方法可以提高参数估计的效率，降低模型的不适定性，保持参数先验信息中的统计、几何或物理意义。

Parameter Estimate Algorithm in Adjustment Model with Interval Constraint

There are usually some uncertain additional information or prior information on parameters in surveying adjustment models,which can constraint on the parameters,and guarantee uniqueness and stability of parameters solution.This paper mainly focused on the studies of adjustment model with box constraints.Firstly,using the regular splitting method of matrix,adjustment model is converted to a simple quadratic programming problem.Then a new iterative algorithm of parameter estimation is established.The algorithm convergence is given.Finally,with example,it is confirmed that the proposed method can improve efficiency of parameter estimation,reduce ill-posed characteristic of model and keep statistical,geometric or physical significance of prior information.