Linear control of time-domain constrained systems

Recent results on the control of linear systems subject to time-domain constraints could only handle the case of closed-loop poles that are situated on the real axis. As most closed-loop systems in practice contain also complex poles, there is a strong need for a general framework encompassing all cases. In this paper such a framework is presented based on sums-of-squares techniques and we show indeed that time-domain constraints on closed-loop signals of linear systems can be incorporated as linear matrix inequalities, even when complex conjugate poles are assigned. The effectiveness of this complete design method is evaluated by means of a simulation example.