Data-Based Modelling for Control
摘要
Control theory has shaped the industrial and technological landscape, from chemical plant control strategies to vehicle automation. Control theory relies on sensor measurements or data from control systems when executing actions. Machine learning can be and is currently used to solve complex control problems, as (i) it provides insights on training, and (ii) it learns from the data. Potentially, machine learning can contribute to optimizing the performance and robustness of systems. In this Chapter, we illustrate the use of machine learning techniques to enhance tuning and optimization capabilities for both the basic control theory and model predictive control.