9 December 2022 in IEEE Transactions on Automatic Control
The problem of establishing out-of-sample bounds for the values of an unknown ground-truth function is considered. Kernels and their associated Hilbert spaces are the main formalism employed herein, along with an observational model where outputs are corrupted by bounded measurement noise. The noise can originate from any compactly supported distribution, and no independent assumptions are made on the available data. In this setting, we show how computing tight, finite-sample uncertainty bounds amounts to solving parametric quadratically constrained linear programs. Next, the properties of our approach are established, and its relationship with another method is studied. Numerical experiments are presented to exemplify how the theory can be applied in various scenarios and to contrast it with other closed-form alternatives.
19 February 2019 in IEEE Transactions on Industry Applications
JPJoao Onofre Pereira Pinto
Joao Onofre Pereira Pinto
10 February 2021 in IEEE Open Journal of the Industrial Electronics Society
Predictive control is a flexible control methodology that can optimize performance while satisfying current and voltage constraints. Its application in the power electronics domain is however hampered by the high computational demands associated with it. In this paper, piecewise-affine neural networks are explored to greatly simplify these controllers and allow for an inexpensive implementation in commercial hardware. More specifically, we tackle the problem of enhancing the start-up transient response of a step-down dc-dc converter while also satisfying inductor current constraints. We analyze the neural network architecture, and detail its training and validation procedures. The learned controller is then embedded on an inexpensive 80-MHz microcontroller, and experimental results are provided showing that the whole control algorithm can be executed in under 30 microseconds.
4 February 2022 in IEEE Transactions on Automatic Control
In this technical note we analyse the performance improvement and optimality properties of the Learning Model Predictive Control (LMPC) strategy for linear deterministic systems. The LMPC framework is a policy iteration scheme where closed-loop trajectories are used to update the control policy for the next execution of the control task. We show that, when a Linear Independence Constraint Qualification (LICQ) condition holds, the LMPC scheme guarantees strict iterative performance improvement and optimality, meaning that the closed-loop cost evaluated over the entire task converges asymptotically to the optimal cost of the infinite-horizon control problem. Compared to previous works this sufficient LICQ condition can be easily checked, it holds for a larger class of systems and it can be used to adaptively select the prediction horizon of the controller, as demonstrated by a numerical example.
1 January 2020 in IFAC-PapersOnLine