Guaranteed robust state feedback controller via constrained optimization using Differential Evolution

Computational intelligence has been successfully applied into many engineering applications including control engineering problems. In this paper, a robust state feedback control design via constrained optimization based on Differential Evolution (DE) is proposed. The feedback controller is designed based on state space model of the plant with structured uncertainty such that the closed-loop system would have maximum stability radius. A wedge region is assigned as a constraint to locate desired closed loop poles. The proposed method is applied into design of anti-swing control of an automatic gantry crane. The experimental result is shown and comparison with that of LQ (linear quadratic) optimal controller is made. The proposed method effectively locates the closed loop poles within the prescribed wedge region and its robust performance is guaranteed.