Institut Français des Sciences et Technologies des Transports, de l'Aménagement et des Réseaux

Type de document

CHAPITRE D'OUVRAGE (CO)

Langue

anglais

Auteur

SCEMAMA

Résumé / Abstract

This paper presents the CLAIRE concepts for a context-free AI based supervisor for traffic control. This system is issued from considerations on adaptive systems, congestion management, and role of operators and the contribution of systems for information and communication with drivers. Claire implements a universal layer which can be interfaced with the specific layer of the control system. It ensures an on-line congestion monitoring and control. It gives off-line aids to the analysis of congestion patterns and a learning capability for the development of congestion control actions. Thus it contributes to the elaboration of a long-term memory of congestion patterns (congestion observatory). Moreover, the conceptual level and architecture of the CLAIRE system allow to address the problem of integration with others advanced road transport systems such as route guidance, traffic information parking management, public transport.