Variational Bayesian Approach for Nonlinear Identification and Control
Matti Tornio and Tapani Raiko
In: IFAC Workshop on IFAC Workshop on Nonlinear Model Predictive Control for Fast Systems (NMPC_FS06), 9-11 Oct 2006, Grenoble, France.
This paper studies the identification and model predictive control in nonlinear state-space models. Nonlinearities are modelled with neural networks and system identification is done with variational Bayesian learning. In addition to the robustness of control, the stochastic approach allows for a novel control scheme called optimistic inference control. We study the speed and accuracy of the two control schemes as well as the effect of changing horizon lengths and initialisation methods using a simulated cart-pole system.