A simplified LMI approach to l/sub 1/ controller design
Anirudh Oberoi, Juan C. Cockburn
Abstract
Anirudh Oberoi, Juan C. Cockburn
Abstract
In this paper a new linear matrix inequality (LMI) for peak-to-peak gain minimization is derived. The new LMI defines a convex upper bound on the non-convex-norm rendering the l/sub 1/-controller design problem convex and facilitating its integration in multiobjective design problems. It also allows the use of projection formulas which simplifies the numerical implementation of the design algorithm.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper a new linear matrix inequality (LMI) for peak-to-peak gain minimization is derived. The new LMI defines a convex upper bound on the non-convex-norm rendering the l/sub 1/-controller design problem convex and facilitating its integration in multiobjective design problems. It also allows the use of projection formulas which simplifies the numerical implementation of the design algorithm.
Key concepts: Linear matrix inequality, Convex optimization, Regular polygon, Norm (philosophy), Mathematics, Minification, Upper and lower bounds, Mathematical optimization