Skip to Main content Skip to Navigation
New interface

Vers des systèmes multi-agents auto-adaptatifs et coopératifs pour les applications d'assistance ambiantes

Abstract : Our research investigates adaptive Multi-Agents Systems (MAS) in open environment. We are interested in an open, dynamic and unstable context in which agents can dynamically cooperate. Since the environment is dynamic, due to the availability or the lack of connected objects and to the diversity of situations, cooperation mechanisms among connected objects must be dynamically and adaptively designed. The adaptation takes into consideration the context of the person (elderly or sick people) as well as the existing interaction protocols.The aim of this work is to propose a contribution to the automation of the design and the deployment of adaptive multi-agent systems for ambient assistive applications. We propose a new approach for Generating Automatically an Adaptive Multiagent system for Ambient Assistive Applications (GAAMAAA). This approach focuses on the design and implementation of the interaction protocol ontology and the creation steps of the MAS. For this purpose, we define the agent-based architecture of the GAAMAAA, we give a protocol ontology for protocol modeling and ambient assistive ontology to create agents. We specify an execution model of protocols that enables agents to dynamically take part into interaction. We carry out the dynamic integration of these protocols into agents involved in the corresponding interaction. Our approach is validated by implementing a localization robot scenario, which the GAAMAAA provides support for two most used interaction protocol ICNP and Coalition. The evaluation shows the effective performance of interaction protocol based on sensitivity analyses method.
Keywords : Multiagent system
Document type :
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Tuesday, July 6, 2021 - 11:01:12 AM
Last modification on : Thursday, January 27, 2022 - 3:04:30 AM
Long-term archiving on: : Thursday, October 7, 2021 - 6:24:21 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03279028, version 1


Amina Maddouri. Vers des systèmes multi-agents auto-adaptatifs et coopératifs pour les applications d'assistance ambiantes. Other [cs.OH]. Université Paris-Saclay; Université de Tunis, 2020. English. ⟨NNT : 2020UPASG035⟩. ⟨tel-03279028⟩



Record views


Files downloads