Speech Title: Learning-based Adaptive Optimal Control and its Applications
Abstract:
Output regulation constitutes a fundamental mathematical framework concerned with designing controllers for dynamic systems to achieve disturbance rejection and asymptotic tracking. Reinforcement learning, on the other hand, provides a paradigm for how an agent can learn optimal strategies through interaction with an unknown or uncertain environment, with the objective of minimizing a cumulative cost over time. Adaptive Dynamic Programming (ADP) emerges as a pivotal branch of reinforcement learning, offering a data-driven, model-free methodology for adaptive optimal control in complex dynamic systems. In this presentation, I will explore the application of ADP as a powerful tool to solve learning-based output regulation problems across both linear and nonlinear systems.
Biography:
Weinan Gao is a Professor, at Northeastern University, IEEE Senior Member, and a Visiting Professor at Mitsubishi Electric Research Laboratories in Boston, USA, and one of the top 2% scientists in the world according to Stanford University. He received his Ph.D. from New York University, USA. He has long been engaged in basic theoretical research on artificial intelligence, adaptive dynamic programming, optimal control and output regulation, and has carried out basic application research on intelligent connected vehicles, autonomous driving, power systems and other specific objects. He has published more than 60 papers in international journals such as IEEE Trans. Automatic Control and Automatica, including 4 ESI highly cited papers. He serves as an editor of international journals in the field of control and artificial intelligence, such as IEEE Journal of Automatica Sinica, IEEE Trans. Neural Networks and Learning Systems, and Control Engineering Practice. He has won several best paper awards at international conferences.