Design of Dynamic Output Feedback Controllers for Linear Systemas Under Constraints
Linear Systems. Invariant Sets. Dynamic Output Feedback. Multiparametric Programming.
This work intends to obtain explicit PWA control laws for systems with constraints by dynamic output feedback. In this regard, the theory of invariant sets and multiparametric programming will be used. Controlled invariant sets have been widely used to solve constrained systems problems. Although it has already been well studied in state feedback control, the use of controlled invariant sets for output feedback still lacks further exploration. The state observer is embedded into the compensator structure, so that there is a dynamic compensator. The proposed output feedback controlled invariant set is constructed from a conditioned invariant set and a controlled invariant set and state uncertainty is reduced using the contraction of the conditioned invariant set. When using linear programming, the control action is computed at each instant of time, with a high computational cost on-line, so the use of multiparametric programming allows to obtain explicit control laws, computed off-line that are stored and used in on-line procedures, thus the computational cost would take place off-line. First, it will be presented the theory of invariant sets applied to the design of controllers using linear programming. Next, the approach using multiparametric programming is presented for cases of state feedback and static output feedback. Finally, it is presented the design of controllers by dynamic output feedback using linear programming and this work proposal controller design using multiparametric programming technique.