On May 23, we organized a scientific seminar titled “Learning to Solve Parametric Mixed-Integer Optimal Control Problems via Differentiable Predictive Control”. Ján Boldocký, a PhD student at the Slovak University of Technology in Bratislava, led the seminar, which was organized within the framework of the FrontSeat project as part of the “Research Seminar on Smart Cybernetics” series.
Abstract:
The presentation introduces a novel approach to solving input- and state-constrained parametric mixed-integer optimal control problems using Differentiable Predictive Control (DPC). The proposed approach follows a differentiable programming paradigm in which an explicit neural policy is established that maps control parameters to integer- and continuous-valued decision variables. The control policy is optimized via stochastic gradient descent by differentiating the quadratic model predictive control objective through the closed-loop finite-horizon response of the system dynamics. The integrality constraints are imposed using three differentiable rounding strategies. The proposed approach is evaluated on a conceptual thermal energy system, comparing its performance with the optimal solution for different lengths of the prediction horizon. The simulation results indicate that the proposed self-supervised learning approach can achieve near-optimal performance while significantly reducing inference time, showing a potential to enable deployment on edge devices.
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).
0 Comments