Geod Sci 762 – Advanced Adjustment Computations U G 4

Professor: Burkhard Schaffrin

Yaron A. Felus

 


Gauss-Markov model and mixed models; collocation technique; generalized matrices in geodetic science; estimable and projected parameters; variance components; prior information; dynamic linear model and Kalman filtering.


Objectives:

The course makes students aware of various special adjustment techniques. Relations between Gauss-Markov model and traditional least squares solutions are explored and compared to the Collocation technique. Ranks of the matrices are discussed, and they are derived for matrices usually encountered in adjustment computations. The introduction to generalized matrices will give the possibility to solve rank-deficient systems. Estimable and nonestimable quantities in adjustment are defined and discussed, as well as the estimation of variance components. The role of prior information is clarified, and it is shown how the least-squares adjustment in a Dynamic Linear Model leads to Kalman filtering.

As a result, students should be able to make a prudent choice of a proper model and the corresponding adjustment techniques for a host of overdetermined problems in geodetic science, no matter how complicated.


 

Class Notes

 

Syllabus:

 

1.     Gauss-Markov Model

1.1   Fundamental theory

1.2   The least squares adjustment

1.3   BLUUE for the parameters

1.4   BIQUUE for the variance component

1.5   BIQE for the variance component

1.6   The E-D correspondence and Kronecker-Zehfuss products of matrices

 

Lab1

 

List of Acronyms and Notations.

 

Reference: Softly unbiased estimation, Part 1: The Gauss-Markov model, Schaffrin, Burkhard

 Linear Algebra and its Applications Vol: 289, Issue: 1-3, pp. 285-296, March 1, 1999

 

Standard Adjustment Models  

 

Lab 2

 

Kronecker - Zehfuss Product formulas

 

2.     Rank-deficient Gauss-Markov model and generalized inverse matrices

2.1   The least-squares adjustment with multiple solutions

2.2   g-inverse matrices and estimable parameters

2.3   Datum constraints

2.4   Restricted least-squares adjustment

 

Lab3

 

Summary of possible g-inverses

 

3.   Variance Components

      3.1   Mixed regression

      3.2   The Variance Components Model

      3.3   Repro-BIQUUE of the variance components

3.4      Unbiasedness of the “adaptive BLUUE”

 

Lab 4

 

4.   Prior information and the Extended Gauss-Morkov Model

      4.1   Pseudo-observations and Wolf’s collocation solution

      4.2   Sequential update of Wolf’s collocation

      4.3   Singular dispersion matrix for pseudo-observations

      4.4   Limit case of datum parameters

4.5     Dual form of the sequential update

 

Models of observation equations (GMM) with Constrains

 

Lab5

 

5.   Prior information and Mixed Model

      5.1   Helmert’s knack and Moritz’s collocation solution

      5.2   Bjerhammar’s modified least-squares principle

5.3     Duality via Helmert’s knack

 

6.   Introduction to the Dynamic Adjustment

      6.1   The Dynamic Linear Model

      6.2   Least-squares adjustment and best linear prediction

      6.3   Kalman filtering, including the dual form

 

Lab 6

 

Links to Kalman Filter

 

Required Textbook:

 

Koch, K. R., Parameter Estimation and Hypothesis Testing in Linear Models, New York: Springer, 1999, 2nd edition.

 

References (for further reading):

Brown, R. G. and P. Y. C. Hwang, Introduction to Random Signals and Applied Kalman Filtering, Wiley: New York, etc. 1997, 3rd edition.

Lütkeohl, H., Handbook of Matrices, Wiley, New York, etc. 1996/97.

Rao, C. R. and H. Toutenburg, Linear Models. Least Squares and Alternatives, New York: Springer, 1999, 2nd edition.

 

Other related books:

 

Gilbert Strang and Kai Borre,  Linear Algebra, Geodesy, and GPS Wellesley-Cambridge Press , 1997  

 

 

WEB sources:

1. SIMPLE STATISTICS & EFFECT STATISTICS