In this paper, we present a new paradigm to construct an optimal map of a multi-parametric quadratic or linear program, based on random sampling. Probabilistic guarantees for coverage of the region over which the map is constructed are provided in terms of user-defined coverage and confidence parameters. Extensive simulations show that the proposed Multi-Parametric Randomized algorithm (MPR) outperforms state-of-the-art competitors.
MPR: A novel randomized algorithm for multi-parametric programming
Falsone, Alessandro;Prandini, Maria
2026-01-01
Abstract
In this paper, we present a new paradigm to construct an optimal map of a multi-parametric quadratic or linear program, based on random sampling. Probabilistic guarantees for coverage of the region over which the map is constructed are provided in terms of user-defined coverage and confidence parameters. Extensive simulations show that the proposed Multi-Parametric Randomized algorithm (MPR) outperforms state-of-the-art competitors.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
MPR_pub.pdf
accesso aperto
:
Publisher’s version
Dimensione
921.73 kB
Formato
Adobe PDF
|
921.73 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
