Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5002886 | IFAC-PapersOnLine | 2016 | 6 Pages |
Abstract
This paper deals with a function block library (blockset) for embedded linear algebra (matrix) computations. The blockset is based on a subset of BLAS and LAPACK subroutines converted to C language with respect to memory limitations of typical embedded platforms. The subset covers algorithms for real matrices with double precision elements in three areas: singular value decomposition, eigenvalues and Sylvester equation. The Implemented blockset contains besides suitably wrapped BLAS and LAPACK algorithms also function blocks for discretization of continuous linear systems and several helper blocks. The typical use of the blockset is demonstrated on examples in the REX control system.
Keywords
Related Topics
Physical Sciences and Engineering
Engineering
Computational Mechanics
Authors
Pavel Balda,