Affiliations: Department of Electrical Engineering, K.N. Toosi
University of Technology, P.O. Box 16315-1355, Tehran, Iran. E-mail:
sedigh@eetd.kntu.ac.ir | Department of Mathematics, MA 4-5, Technical
University of Berlin, P.O. Box 10623, Berlin, Germany. E-mail:
ybavafat@math.tu-berlin.de
Note: [] Corresponding author
Abstract: New necessary and sufficient conditions for multivariable pole
placement (MVPP) and entire eigenstructure assignment (EEA) through static
linear multivariable output feedback are established. It is shown that the
resultant matrix is of full rank and all design freedoms are retained. The
problem of static linear multivariable output feedback control law design is
then defined. Based on the EEA concept and sufficiency of the regional pole
placement, the design is (re)formulated in terms of a constrained nonlinear
optimization problem. To this end, some decoupling indices for noninteractive
performance are defined, their necessary and sufficient conditions are derived
and tracker design is addressed. The problem formulation well suits the
application of random/intelligent optimization techniques. By way of this
approach, optimal robust stability/performance, noninteractive performance,
reliability, actuator limitations and low sensitivity in the face of structured
or unstructured plant uncertainties are achieved. The effectiveness of the
proposed methodology is demonstrated by simulation results using genetic
algorithm.
