Multi-objective switching control via LMI optimization

Multi-objective design problem is the optimization of various and often conflicting objectives for a complex system. In this paper, optimization is performed using Linear Matrix Inequalities (LMI's). A switching strategy is proposed in order to improve the multi-objective control performance. Each controller design is based on a set of performance specifications. Instead of considering all the specifications defined by respective LMI sets simultaneously, only relevant objectives are included in the control design procedure. Then, switching is performed to meet multiple objectives. Assurance of the overall stability of the closed-loop is acknowledged via specific controller realization. Multi-objective designs are prone to conservatism, which is greatly reduced by the switching approach. The efficiency of the proposed methodology is illustrated through an example.