Title :
A Hamiltonian-Jacobi algorithm
Author_Institution :
Dept. of Math., Kansas Univ., Lawrence, KS, USA
fDate :
5/1/1990 12:00:00 AM
Abstract :
The nonsymmetric Jacobi iteration is adapted to the special structure of Hamiltonian matrices. This Hamiltonian-Jacobi algorithm uses symplectic-unitary similarity transformations to solve algebraic Riccati equations through the Hamiltonian-Schur form. It preserves Hamiltonian structure without using a condensed form. Although it converges too slowly for use on conventional serial computers, it may be attractive for some highly parallel architectures
Keywords :
iterative methods; matrix algebra; Hamiltonian matrices; Hamiltonian-Jacobi algorithm; Hamiltonian-Schur form; algebraic Riccati equations; nonsymmetric Jacobi iteration; symplectic-unitary similarity transformations; Concurrent computing; Control theory; Eigenvalues and eigenfunctions; Jacobian matrices; Mathematics; Matrix decomposition; Numerical stability; Parallel architectures; Riccati equations; Roundoff errors;
Journal_Title :
Automatic Control, IEEE Transactions on
