附有病态约束反演问题的岭估计法

提出了处理附有病态约束矩阵的等式约束反演问题的岭估计解法,推导了相应的求解公式及均方误差评定精度的方法,给出了确定附有病态约束矩阵的等式约束反演中岭参数的岭迹法、广义交叉核实法和L曲线法。算例比较了三种确定岭参数的方法在处理附有病态约束矩阵的等式约束反演问题中的优缺点,并与截断SVD法进行了对比分析。     

采用联合平差法处理附有病态等式约束的反演问题

探讨了附有病态等式约束的反演问题,尝试用降秩处理方法解这类病态约束问题,通过算例验证了此种方法与截断奇异值方法是等价的。然后提出了一种联合平差方法,它不仅能解病态的约束问题,而且能解决主模型秩亏或病态同时约束模型病态的问题,增强了应用性。最后设计了多种方案进行计算和比较,验证了联合平差法的有效性和可行性。     

附有病态等式约束反演问题的联合岭估计

基于非等精度观测探讨了附有病态等式约束的反演问题;针对模型中的系数矩阵和约束矩阵同时存在病态性的情况,提出了联合岭估计,并推导了岭参数的确定方法。理论分析和实验结果均表明,联合岭估计不仅能够消除约束矩阵病态的不良影响,而且能较好地克服主模型系数矩阵病态和约束矩阵病态同时存在而产生的不稳定性,具有良好的性质。     

等式约束EIV模型的Newton-Gauss迭代解法及其精度评定

利用平差参数间的合理等式约束能够提高解的稳定性。针对变量误差模型EIV(errors-in-variables)引入等式约束,分别针对系数阵良态和病态两种情形建立了约束总体最小二乘准则。基于非线性最小二乘问题的常用解法Newton-Gauss法,由约束准则构建了拉格朗日极值函数并由欧拉-拉格朗日必要条件导出了等式约束EIV模型的Newton-Gauss迭代解。针对精度评定时未考虑参数估值偏差所带来的影响这一不足,基于蒙特卡罗模拟法提出了一种估计约束EIV模型单位权方差和参数估值的协方差阵的数值方法。算例分析结果表明,约束总体最小二乘解严格满足先验等式约束条件;当系数阵病态时,约束条件能够提升解的稳定性和精度。此外,基于蒙特卡罗的数值方法能够获得稳定且合理的精度评定结果。     

不等式约束卡尔曼滤波的解算及其统计性质

卡尔曼滤波是研究如何从被噪声污染的观测信号中过滤噪声,尽可能消除噪声影响,求未知真实信号或系统状态的一种估计方法.首先简要回顾了无约束离散时间不变系统卡尔曼滤波的模型及解算方法、统计性质.然后将其扩展到等式约束情形,推导了等式约束卡尔曼滤波的解及其统计性质.根据有效约束集的思想阐明了附不等式约束和等式约束卡尔曼滤波问题的内在联系,指出其解具有相同的性质,并提出用积极集法解决具有二次规划形式的不等式约束卡尔曼滤波问题.     

基于有效约束的附不等式约束平差的一种新算法

不等式约束是客观实际中普遍存在的一种约束,但目前大地测量数据处理领域并没有成熟、完整并被普遍接受的处理理论和方法。首先简要总结附不等式约束平差的各种方法及其存在的问题。然后对现有测量平差中附有等式约束的平差模型进行扩展,提出一种新的处理附有线性约束(包括等式和不等式约束)的平差方法。该方法在有效约束概念下,通过库恩-塔克条件来寻找有效约束条件,把不等式约束平差问题转化为我们熟知的等式约束平差问题,因此实现解向量与观测向量之间的显式表达。最后,用一个数值算例验证新方法的可行性,同时算例分析表明:用等式约束代替有效约束或集成约束进行平差计算,能得到正确的平差结果,但得不到正确的精度评定结果。     

等式约束病态模型的截断奇异值解及其统计性质

利用平差模型间合理的先验信息能够显著提高解的稳定性和精度。本文基于病态模型引入等式约束条件,并采用截断奇异值法重构了系数阵以削弱其病态性;建立了修正等式约束模型,导出了病态模型的约束截断奇异值解及其偏差、方差以及均方误差公式;分析了截掉奇异值所引起解的偏差引入量与方差下降量的关系,得到了确定截断参数的条件。数值算例和病态测边网算例分析结果表明,最小二乘解严重偏离真值,500次模拟实验的平均RMSE为6.693 5,正则化解和截断奇异值解精度较最小二乘解有所提高,平均RMSE分别为0.365 8和0.365 2;本文提出的约束截断奇异值解的精度最高,与约束正则化解精度相当,其平均RMSE仅为0.057 3。     

附有病态约束矩阵的等式约束反演问题研究

讨论等式约束反演问题的求解方法,给出约束矩阵的病态性的定义及相应的评价方法,推导约束矩阵病态对反演结果的影响公式,通过简单的例子说明约束矩阵病态对反演结果的影响;采用处理约束矩阵病态的SVD方法.有效解决约束矩阵病态造成反演结果极不稳定的问题.最后设计多个反演方案进行计算,验证方法的有效性和可行性.     

附加等式约束的病态模型谱修正迭代法

受法矩阵小奇异值的影响,病态模型的最小二乘估值极不可靠,常用的正则化法如Tikhonov正则化法、截断奇异值法等均为有偏估计,谱修正迭代法通过在法方程两端加上改正项纠正了方程的病态性,反复迭代得到了病态方程的稳定数值解,因其未改变方程的等式结构,因此是一种无偏估计。本文在一般谱修正迭代法的基础上引入等式约束,利用Lagrange乘数法导出了等式约束病态模型的谱修正迭代解,并通过数值算例验证了公式的有效性和可行性。     

利用不等式约束求解病态问题的新算法

病态问题是大地测量数据处理中常见的问题,充分利用平差过程所给的先验信息可以确保参数的可靠性和有效性.提出了一种利用不等式约束求解病态问题的新算法,该算法将先验信息表示为不等式形式,并与病态模型构成不等式约束平差模型.结合Karush-Kuhn-Tucker条件可将该模型转化为线性互补问题,然后利用Lemke算法求解.该法避免了对病态矩阵求逆,保证了参数解的唯一性和稳定性.最后,本文模拟了未知参数附先验信息的Hilbert矩阵及全球定位系统(Global Positioning System,GPS)快速定位实验,并结合多种经典的病态平差方法,验证了Lemke算法在处理病态问题上的有效性.     

An Aggregate Constraint Method for Inequality-constrained Least Squares Problems

The inequality-constrained least squares (ICLS) problem can be solved by the simplex algorithm of quadratic programming. The ICLS problem may also be reformulated as a Bayesian problem and solved by using the Bayesian principle. This paper proposes using the aggregate constraint method of non-linear programming to solve the ICLS problem by converting many inequality constraints into one equality constraint, which is a basic augmented Lagrangean algorithm for deriving the solution to equality-constrained non-linear programming problems. Since the new approach finds the active constraints, we can derive the approximate algorithm-dependent statistical properties of the solution. As a result, some conclusions about the superiority of the estimator can be approximately made. Two simulated examples are given to show how to compute the approximate statistical properties and to show that the reasonable inequality constraints can improve the results of geodetic network with an ill-conditioned normal matrix.     

A solution to EIV model with inequality constraints and its geodetic applications

In the field of surveying, mapping and geodesy, there have been a number of works on the error-in-variable (EIV) model with constraints, where equality constraints are imposed on the parameter vector. However, in some cases, these constraints may be inequalities. The EIV model with inequality constraints has not been fully investigated. Therefore, this paper presents an inequality-constrained total least squares (ICTLS) solution for the EIV model with inequality constraints (denoted as ICEIV). Employing the proposed ICTLS method, the ICEIV problem is first converted into an equality-constrained problem by distinguishing the active constraints through exhaustive searching, and it is then resolved employing the method of equality-constrained total least squares (ECTLS). The applicability and feasibility of the proposed method is illustrated in two examples.     

NLS问题的不适定性及正则化解算方法

本文进一步完善定义了NLS问题的两种不适定性,对产生这两种不适定问题的现象进行了分析。借助于正则化理论,通过添加稳定泛函,结合高斯-牛顿法,构造了不适定NLS问题的正则化高斯-牛顿法求解公式;解决了普通高斯-牛顿法在迭代过程中其Jacobian矩阵是秩亏或者严重病态导致的不能收敛的问题;给出了非线性秩亏自由网平差的正则化高斯-牛顿法步骤;以几个经典NLS问题为例进行了数值实验,说明了本文所提方法的适用性。     

火山形变Mogi模型反演的病态总体最小二乘解算方法

相对最小二乘方法,总体最小二乘顾及了观测方程系数矩阵含有误差的情况,然而,当系统出现病态时,总体最小二乘受病态的影响将更加明显。因此,针对病态总体最小二乘问题解算方法的研究越来越多受到关注。文中基于总体最小二乘进行火山形变Mogi模型反演,针对反演过程中出现的病态性问题,采用虚拟观测解法、谱修正迭代解法、共轭梯度解法,通过模拟算例验证文中方法在抑制病态性方面的有效性。与一般总体最小二乘、正则化总体最小二乘等方法相比存在优势。     

遥感反演中约束最优化方法的拓展

遥感反演大多是典型的约束是最优化问题。本文对现有的约束最优化方法在遥感反演中的适用性进行了分析，从提高反演速度有降低优化方法病态特性两个方面考虑，提出了罚矩阵的概念，对约束优化方法中的乘子法进行了拓展，并进行了理论证明。经过大量的模拟反演实验表明，拓展后的乘子法的反演速度提高了大约30％，病态程度也得到了进一步的改善。     

等式约束的效率分析和一致性检验

利用正定二次型矩阵极值的方法进一步分析等式约束对平差结果的影响,推导出等式约束线性模型的验后单位权方差(中误差)最小值,验证了该值和无等式约束的方差一致性,并将等式约束线性模型的验后单位权中误差最小值作为等式约束的验后单位权中误差限差的基准值。     

Total least squares adjustment in partial errors-in-variables models: algorithm and statistical analysis

The weighted total least squares (TLS) method has been developed to deal with observation equations, which are functions of both unknown parameters of interest and other measured data contaminated with random errors. Such an observation model is well known as an errors-in-variables (EIV) model and almost always solved as a nonlinear equality-constrained adjustment problem. We reformulate it as a nonlinear adjustment model without constraints and further extend it to a partial EIV model, in which not all the elements of the design matrix are random. As a result, the total number of unknowns in the normal equations has been significantly reduced. We derive a set of formulae for algorithmic implementation to numerically estimate the unknown model parameters. Since little statistical results about the TLS estimator in the case of finite samples are available, we investigate the statistical consequences of nonlinearity on the nonlinear TLS estimate, including the first order approximation of accuracy, nonlinear confidence region and bias of the nonlinear TLS estimate, and use the bias-corrected residuals to estimate the variance of unit weight.     

On weighted total least-squares with linear and quadratic constraints

A weighted total least-squares (WTLS) approach with linear and quadratic constraints is developed. This method is according to the traditional Lagrange approach to optimize the target function of this problem. The WTLS and constrained total least-squares (CTLS) approach had been distinctively investigated, however, these two problems have not been simultaneously considered yet; furthermore, among the contributions on the CTLS problem, only Schaffrin and Felus considered linear and quadratic constraints together; nevertheless, in many practical examples, some elements of the design/coefficient matrix are fixed and should not be modified and this approach cannot deal with these cases. The main necessity of this research appears after the desirable property of the WTLS approach in preserving the structure of the design matrix was proven by Mahboub. In other words, currently, the WTLS approach is one of the most efficient methods for solving the so-called errors-in-variables model and an attempt for equipping it with constraints seems necessary. Also it is demonstrated that the additional constraints have a ’regularization role’ for ill-conditioned problems and the unconstrained solution suffers from ill-conditioning effects which give it an added advantage over the unconstrained WTLS algorithm. Four geodetic applications indicate the significant of this problem in the presence of colored and white noise in the data.     

修正岭估计方法在测量数据处理中的应用研究

在测量数据处理中,若法方程呈现病态,岭估计可以改善方程的病态性,提高解的精度,使得在均方根意义下,岭估计解优于最小二乘解;另外,观测对象的先验信息也可以用来改善解的效果.文中借鉴一种修正岭估计方法,将岭估计方法和先验信息联系起来,并以GPS数据解算为例,将修正岭估计方法与普通解法进行比较,说明这种应用方法的合理性.