ROBUST-CONTROL OF CONSTRAINED SYSTEMS VIA CONVEX-OPTIMIZATION

A successful controller design paradigm must take into account both mo del uncertainty and design specifications. Model uncertainty can be su ccessfully addressed using H-infinity robust control theory. However, this framework cannot directly accommodate the realistic case where in addition to robustness considerations the system is subject to both t ime- and frequency-domain specifications, such as bounds on the contro l action. In this paper we propose a design procedure, based upon the use of convex optimization, that takes explicitly into account both ti me- and frequency-domain specifications. The main result of the paper is a new framework to address problems having both control and output constraints and model uncertainty. Additionally, the paper serves as a brief tutorial on the issues involved in addressing design problems w ith multiple design specifications via convex optimization.