Mixture-based adaptive probabilistic control

摘要

Quasi-Bayes algorithm, combined with stabilized forgetting, provides a tool for efficient recursive estimation of dynamic probabilistic mixture models. They can be interpreted either as models of closed-loop with switching modes and controllers or as a universal approximation of a wide class of non-linear control loops. Fully probabilistic control design extended to mixture models makes basis of a powerful class of adaptive controllers based on the receding-horizon certainty equivalence strategy. Paper summarizes the basic elements mentioned above, classifies possible types of control problems and provides solution of the key one referred to as 'simultaneous' design. Results are illustrated on mixtures with components formed by normal auto-regression models with external variable (ARX).