A one-day workshop called Group of Global Excellence (GoGE) was held on last October 16. This workshop was organized by CDSL and supported by Brain Korea Program of the Korean government.
Professor Dragan Nesic of the Department of Electrical and Electronic Engineering at University of Melbourne, Australia, was invited as the main speaker of this workshop. Prof. Nesic delivered a seminar under the title of “Regularzing policy iteration for recursive feasibility and stability”. In this seminar, he discussed how to fill a gap between optimal control theory and stability theory: stability is often overlooked in optimal control theory and optimality is often missed in stability theory. Buliding upon the policy iteration (PI) method, which is an widely used classical method of optimal control, prof. Nesic proposed the PI+ algorithm by adding policy regularization to PI to ensure the recursive feasibility. He also showed that the controller created by PI+ ensures near-optimality, closed-loop stability, and even robust stability. This was a good inspiration for the seminar participants.
Several students of CDSL participated as speakers in this workshop. Joowon Lee shared her recent result under the title of “Conversion of controllers to have integer state matrix for encrypted control: Non-minimal order approach”, Hamin Chang shared “Active inverse identification by Gaussian process regression and its application to control”, and Soojeong Hyeon shared “Multi-agent target position estimation using bearing-only measurements via spatial excitation”.
Overall, this seminar was a great time for participants to learn and to share creative ideas about control theory.