Servicio de Información Comunitario sobre Investigación y Desarrollo - CORDIS


MPA Informe resumido

Project ID: G1RD-CT-2000-00298
Financiado con arreglo a: FP5-GROWTH
País: Hungary

Method to improve multi-agent based scheduling by neurodynamic programming

This result is a new machine learning approach developed by SZTAKI that may be generally applied in the control software of the MPA (Modular Plant Architecture) Software Tool. Applying reinforcement learning, the new approach improves the decision-making in the control software. A two level adaptation method is proposed to solve the scheduling problem in a dynamically changing and uncertain environment. It is applied to the heterarchical multi agent architecture of the MPA controller, which was inspired by food forging ants. The applicability and the effectiveness of the proposed method are illustrated by the results of experimental runs.

Reported by

Computer and Automation Research Institute, Hungarian Academy of Sciences, SZTAKI
Kende u. 13-17
1111 Budapest
See on map
Síganos en: RSS Facebook Twitter YouTube Gestionado por la Oficina de Publicaciones de la UE Arriba