On October 25, RUB Bochum organized a scientific seminar on “Explicit Model Predictive Control for PDEs”. The scientific seminar was led by Mikael Kurula, a senior university lecturer at Mathematics in the Faculty of Science and Engineering, Åbo Akademi University. The scientific seminar was organized in the framework of the FrontSeat project and held in a hybrid format, i.e., as in-person and online lecture.
Abstract: Explicit model predictive control design is carefully developed for linear plants on Hilbert spaces, and we highlight the role of the so-called Slater condition in the reliable explicit solution of the MPC optimization. We then proceed to present a region-free active-set algorithm for explicit MPC, using it to stabilize an unstable, constrained heat equation with Neumann boundary control. The example demonstrates the ability to deal with a high prediction model dimension and a long optimization horizon. We briefly look at a more elaborate version of the active-set algorithm that is guaranteed to terminate in finite time in the absence of state constraints. We end with an application of the discrete-time explicit MPC controller to an unstable Timoshenko beam equation operating in continuous time. Throughout the talk, opportunities for further development will be pointed out.
This project has received funding from the European Union’s Horizon under grant no. 101079342 (Fostering Opportunities Towards Slovak Excellence in Advanced Control for Smart Industries).