A Riccati-based FH norm optimization algorithm for robust low-order controller design

A computational algorithm is presented for determining the optimal controller in the Frobenius-Hankel (FH) norm approach for control of discrete-time systems. The algorithm is based on the use of algebraic Riccati equations which in the limit reduce to Lyapunov equations that appear in the necessary conditions. It is shown that the FH norm approach provides computational ease for a broad class of problems and a near-optimal solution to the H/sub infinity / norm minimization for controllers of bounded order. The Riccati equation-based algorithm is computationally attractive since it takes advantage of computational tools developed for the Riccati equation and eliminates the search for an initial stabilizing solution.