On-Line Tunable Explicit Model Predictive Control for Systems with a Fast Dynamics

Period: 1.1.2019 - 31.12.2022

Project Code: VEGA 1/0585/19

Principal investigator: Michal Kvasnica

Scientific co-workers: Martin Klaučo, Miroslav Fikar, Juraj Oravec, Peter Bakaráč

Aims of the project: The aim of the project is the development of a unified methodology for the design, synthesis, and implementation of explicit model predictive controllers that can be tuned on-line by changing the parameters of the cost function and/or of the prediction model. Explicit predictive controllers are known to combine quality and safety of nonlinear control algorithms with the cheap implementation complexity known from linear controllers. Therefore they allow for an optimal and safe regulation of systems with a fast dynamics with time constants in the order of milli- to micro-seconds. Their main drawback, however, is that they cannot be re-tuned on-line. Mitigation of this drawback will lead to extension of the current knowledge in the areas of optimal and predictive control and, more importantly, will enable such controllers to be employed in process automation where quality and safety of control algorithms is of paramount importance. 



  1. M. KvasnicaP. BakaráčM. Klaučo: Complexity reduction in explicit MPC: A reachability approach. Systems & Control Letters, vol. 124, pp. 19–26, 2019.
  2. J. Oravec – J. Holaza – M. Horváthová – N. A. Nguyen – M. KvasnicaM. Bakošová: Convex-lifting-based robust control design using the tunable robust invariant sets. European Journal of Control, 2019.

Responsibility for content: Ing. MSc. Martin Klaučo, PhD.
Last update: 21.12.2018 12:45
Facebook / Youtube

Facebook / Youtube