Data-driven control design for nonlinear systems

Data-driven control has emerged as an important field within control theory, which represents a foundational step in bridging machine learning and control systems. Although extensive research has been conducted over the years, most available results center on linear systems, with limited attention given to nonlinear cases. For unknown nonlinear systems with available data, synthesizing controllers from data with rigorous theoretical guarantees becomes essential. 

In his thesis, Zhongjie Hu proposes novel control design methods to learn controllers from data for nonlinear systems. Firstly, Hu proposes a data-driven control design method for nonlinear systems that builds on kernel-based interpolation. Secondly, he presents data-based conditions for enforcing contractivity via feedback control and obtain desired asymptotic properties of the closed-loop system. Finally, Hu presents the data-based design of state-feedback controllers that solve the output regulation problem for a class of nonlinear systems.

Dissertation: https://hdl.handle.net/11370/2f1e9360-04b8-41d8-aadf-f5984969fa5e