Search

Based on a proposed complete parametric approach for eigenstructure assignment in multivariable linear systems via dynamical compensators, insightful parametrizations of the closed loop eigenvalues sensitivities to the perturbed elements in the open loop system matrices are obtained, and an effective algorithm for eigenvalue assignment with minimum sensitivity in multivariable linear systems via dynamical compensators is then proposed. The algorithm does not contain ‘going back’ procedures, and allows the closed loop eigenvalues to be conveniently optimized within desired regions. A numerical example demonstrates its effect, simplicity and numerical property.

In this paper we examine the problem of robust pole placement using state-feedback in generalized systems. We develop a robustness theory for the finite generalized spectrum of the system as a partial problem, and for the “infinite” pole placement problem as a second partial problem where perfect conditioning is always achievable. We also give bounds between the complete closed-loop robust eigenvalue problem and the two partial ones. The basic tool that is exploited in the presented theory, is the concept of chordal metric. In this paper we take advantage of this notion and present a compact theory for the robust eigenvalue assignment problem in generalized systems. The proposed theory is easy to implement, retrieves the results for the state-variable case as a special case, and takes advantage of well-known computational results.

A design procedure for decentralized controllers based on eigenspectrum assignment for a large-scale system composed of symmetrically interconnected subsystems is presented. The procedure is executed at the subsystems level. All subsystems share the global assignment process.

This paper will investigate the controllability properties for systems parameterized as in Harns et al. (1983, Automatica , 19, 551–555). It will be shown that for systems of low dimensions this parameterization must be done carefully to guarantee that the system is controllable over the parameterization. The controllability property is key to the algorithms developed in Lefebvre et al. (1985, Int. J. Control , 41 , 1273–1292), Richter and DeCarlo (1984, IEEE Trans. Aut. Control , AC-29 ), Harris et al. (1983, Automatica , 19 , 551–555), Richter et al. (1981, Proc. 1981 Joint Aut. Control Conf.), Harris and DeCarlo (1982, Purdue School of EE, TR-EE 82-11) and Sebok and DeCarlo (1984, Proc. Allerton Conf. Communication, Control and Computing).