| 1. |
- Niu, Steve S., et al.
(författare)
-
A Tutorial On Multiple Model Least-Squares and Augmented UD Identification
- 1995
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- The augmented UD identification (AUDI) is a family of new identification algorithms that are based on some well-known matrix decomposition and updating techniques. Compared with conventional least-squares methods, the AUDI methods are conceptually more concise, computationally more efficient, numerically more robust and application-wise more complete. As a result, AUDI is recommended as a complete replacements for conventional recursive least-squares in all parameter estimation and system identification applications. This tutorial paper presents an overview of the AUDI concept, implementation and applications. The multiple model least-squares (MMLS) method, which is a fundamental reformulation and an efficient implementation of the basic least-squares method, is discussed first. Some application examples of the MMLS/AUDI are presented to demonstrate the versatility and reliability of this type of algorithms.
|
|
| 2. |
- Niu, Steve S., et al.
(författare)
-
Decomposition Methods for Solving Least-Squares Parameter Estimation
- 1995
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- In this paper leastsquares method with matrix decomposition is revisited and a multiple model formulation is proposed The proposed formulation takes advantage of the wellestablished decomposition methods but possesses a multiple model structure which leads to simpler and more exible implementations and produces more infor mation than the original least squares methods Several application examples in signal processing and system identication are included As a basic numerical analysis tool the proposed methods can be used in many dierent application areas
|
|
| 3. |
- Niu, Steve S., et al.
(författare)
-
Decomposition Methods for Solving Least-Squares Parameter Estimation
- 1996
-
Ingår i: IEEE Transactions on Signal Processing. - : Institute of Electrical and Electronics Engineers (IEEE). - 1053-587X .- 1941-0476. ; 44:11, s. 2847-2852
-
Tidskriftsartikel (refereegranskat)abstract
- A multiple model least-squares method based on matrix decomposition is proposed. Compared with the conventional implementation of the least-squares method, the proposed method is simpler and more flexible in implementation and produces more information. An application example in parameter estimation is included. As a basic numerical tool, the proposed method can be used in many different application areas.
|
|
| 4. |
- Niu, Steve S., et al.
(författare)
-
Hinging Hyperplanes for Non-Linear Identification
- 1995
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- The hinging hyperplane method is an elegant and efficient way of identifying piecewise linear models based on the data collected from an unknown linear or nonlinear system. This approach provides "a powerful and efficient alternative to neural networks with computing times several orders of magnitude less than fitting neural networks with a comparable number of parameters", as stated in [3]. In this report, the hinging hyperplane approach is discussed from the system identification viewpoint. The bottleneck of this approach, namely, the hinge finding scheme, is investigated. The behavior of the hinge finding algorithm is very dependent on the initial values provided. Several methods for analyzing low dimensional cases are suggested. Although not general, these methods provide some interesting insights into the mechanisms of the hinge finding algorithm. Information from linear models produced by the multiple model least-squares is used to facilitate implementation. The possibility of using binary-tree structured models is also discussed. In addition, an extension of the hinging hyperplane idea leads to a hinge smoothing method in which the hinging hyperplanes are smoothed at the hinge. As a result a neural net like basis function is obtained. Finally, the hinging hyperplane method is used for modeling three real systems.
|
|
| 5. |
|
|
