A Fully Probabilistic Design for Tracking Control for Stochastic Systems With Input Delay

This article studies model reference adaptive control (MRAC) for a class of stochastic discrete time systems with delays in the input. In particular, unified fully probabilistic framework is established to develop solution MRAC, where controller minimizer Kullback-Leibler divergence between actual and desired joint probability density functions tracking error controller. The developed quite general, all components within this framework, including system error, are modeled using models. general arbitrary models first obtained then demonstrated on linear Gaussian delay main input, thus obtaining results. contribution twofold. First, we design referred as design, dynamical systems. Second, establish systematic pedagogic procedure that based deriving explicit forms required predictive distributions causal form randomized when input present.