Hauser, Jan (Czech Technical University in Prague), Pachner, Daniel (Honeywell), Havlena, Vladimir (Honeywell Intl.)

Gaussian Process Based Model-Free Control with Q-Learning

Scheduled for presentation during the Regular Session "Learning and Control" (ThB2SP), Thursday, August 22, 2019, 17:00−17:20,

5th IFAC International Conference on Intelligent Control and Automation Sciences, August 21-23, 2019, Queen’s University Belfast, Northern Ireland

This information is tentative and subject to change. Compiled on April 15, 2024

Abstract

The aim of this paper is to demonstrate a new algorithm for Machine Learning (ML) based on Gaussian Process Regression (GPR) and how it can be used as a practical control design technique. An optimized control law for a nonlinear process is found directly by training the algorithm on noisy data collected from the process when controlled by a sub-optimal controller. A simplied nonlinear Fan Coil Unit (FCU) model is used as an example for which the fan speed control is designed using the o-policy Q-learning algorithm. Additionally, the algorithm properties are discussed, i.e. learning process robustness, GP kernel functions choice. The simulation results are compared to a simple PI, designed based on a linearized model.