A quadratic programming approach for solving the l/sub 1/ multiblock problem (Q4507201)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1518702
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A quadratic programming approach for solving the l/sub 1/ multiblock problem |
scientific article; zbMATH DE number 1518702 |
Statements
A quadratic programming approach for solving the l/sub 1/ multiblock problem (English)
0 references
17 October 2000
0 references
computational methods
0 references
\(Q\)-design
0 references
multiblock \(l_1\)-problem
0 references
bounds
0 references
convergence
0 references
semidefinite quadratic programming
0 references
0.8511093258857727
0 references
0.8461364507675171
0 references
0.7768747806549072
0 references
0.7650210857391357
0 references
0.7538071274757385
0 references
The general multiblock \(l_1\)-problem is equivalent to an infinite-dimensional linear program. Several methods are known to provide approximate solutions to this problem. In this paper a new method is given to compute suboptimal solutions to the standard \(l_1\)-problem. This method provides converging upper and lower bounds to the optimal cost and convergence in norm to the optimal \(l_1\)-solution. The norm convergence is guaranteed independently of the input and output order. The suboptimal \(Q\)-parameter is directly computed and it is easy to obtain the corresponding controller. One has (for each \(N\)) to solve a semidefinite quadratic programming problem instead of a linear one. It seems to be possible to extend the method to compute solutions to other mixed objective problems.
0 references
