Stochastic feedforward—feedback control design in a discrete-time case

Abstract

This paper examines the role of random feedforward inputs in a probabilistic framework for stochastic control design in a discrete-time case. Fundamental rules of probability are applied to extend the Chapman-Kolmogorov equation by including the effect of the additional random disturbance variables. The steady state solution to this equation is then applied to the design of two long-term regulatory controllers: a probability-density-shaping controller and an approximation of an optimal controller with respect to a non-symmetric and non-quadratic cost functions.