×

Newton’s approach to gain-controlled robust pole placement. (English) Zbl 0890.93040

A computational method for robust pole placement of multi-input linear control systems is proposed. Here, the feedback matrix is obtained by minimizing simultaneously the sensitivity of the closed loop poles and the norm of the feedback matrix. Analytical expressions for the first and second derivatives of the cost function are derived which allows to use standard unconstrained minimization routines based on the Newton’s method in finding the optimal gain matrix. The basic limitation of the method is the requirement that the desired poles are different from the open-loop poles of the system. A numerical example for pole placement of a fifth-order two-input system is presented to illustrate the proposed technique in detail and comparisons with other existing methods are made.

MSC:

93B55 Pole and zero placement problems
93B40 Computational methods in systems theory (MSC2010)
93B35 Sensitivity (robustness)
PDFBibTeX XMLCite
Full Text: DOI