PROJECT TITLE :

Stochastic Control for Linear Systems With Additive Cauchy Noises

ABSTRACT:

An optimal predictive controller for linear, vector-state dynamic systems driven by Cauchy measurement and process noises is developed. For the vector-state system, only the characteristic function of the conditional likelihood density operate (pdf), and not the pdf itself, will be expressed analytically in a closed kind. Consequently, the conditional performance index formulated for the controller design can additionally be evaluated only in the spectral domain. In particular, by taking the conditional expectation of an objective perform that is a product of functions resembling Cauchy pdfs, the conditional performance index is obtained in closed form by using Parseval's identity and integrating over the spectral vector. This forms a deterministic, non-convex function of the control signal and therefore the measurement history that must be optimized numerically at each time step. A two-state example is used to expose the interesting robustness characteristics of the proposed controller.