Comparing Disjunctive and Concise Approaches for Set-Guaranteed Estimation in Max-Plus Linear Systems
Résumé
This study compares an existing method with a novel approach for state estimation of Max-Plus Linear systems with bounded uncertainties. Traditional stochastic filtering does not apply to this system class, despite computable posterior probability density function (PDF) support. Existing literature suggests a limited scalability disjunctive approach using difference-bound matrices. To overcome this, we study an alternative method recently investigated in Mufid et al. (2022) using Satisfiability Modulo Theory (SMT) techniques, which are known to be NP-hard. We propose a concise method that utilizes a pseudo-polynomial time algorithm using max-plus algebra. We evaluate its efficiency against SMT techniques through numerical experiments involving sparse matrix multiplications for enhanced computational speed.
Origine | Fichiers éditeurs autorisés sur une archive ouverte |
---|