Recently, a novel method for feedback control of linear systems called "Generalized Sampled-Data Hold Function Control" has received attention. This method uses extensively the power of microprocessors. The original feature of the method, as compared to conventional sampled data control, is that one of the design variables is the hold function itself, rather than just the coefficients of the compensator. A preliminary study has shown advantages of this method as compared to dynamic compensation: it has the efficacy of state feedback without requiring state estimation, it easily allows simultaneous controller design, it requires few on-line computations, but uses only computer memory. The purpose of this research is threefold. The PI intends to achieve a better understanding of the use of Generalized Sampled Data Hold Functions by solving a greater variety of problems which are specific to this method, to formulate and quantify all the design tradeoffs involved, and to develop rules of thumb for use in the practical implementation of this method. Among specific problems to investigate are: inter-sampling behavior, period optimization, stability margin optimization, simultaneous stabilization, robust synthesis, tracking and disturbance rejection, amplitude limitations, and nonlinear system stabilization.
