{"id":315,"date":"2022-12-09T15:16:25","date_gmt":"2022-12-09T14:16:25","guid":{"rendered":"https:\/\/www.mines-stetienne.fr\/fayol\/?page_id=315"},"modified":"2022-12-09T15:16:27","modified_gmt":"2022-12-09T14:16:27","slug":"these-preparee-par-iago-felipe-trentin","status":"publish","type":"page","link":"https:\/\/www.mines-stetienne.fr\/fayol\/theses\/these-preparee-par-iago-felipe-trentin\/","title":{"rendered":"Th\u00e8se pr\u00e9par\u00e9e par Iago Felipe Trentin"},"content":{"rendered":"\n

Titre : Adaptation d’Environnements Intelligents Centr\u00e9e sur l’Humain : une Approche de Raisonnement Multi-agents et Bas\u00e9e sur les Valeurs<\/strong><\/p>\n

Title : Human-centered Adaptation of Smart Environments: a Multi-agent and Value-based Reasoning Approach<\/strong><\/p>\n

Beginning of thesis :<\/strong> 2018
End of thesis :<\/strong> 2021<\/p>\n

R\u00e9sum\u00e9 : <\/strong>Les espaces sociaux \u00e9quip\u00e9s d’appareils de l’Internet des Objets (IdO) sont plus courants. Ces environnements intelligents sont possibles gr\u00e2ce \u00e0 l’Intelligence Artificielle (IA), et leur objectif est de rendre la vie des habitants plus confortable. Ainsi, les humains sont un aspect central des environnements intelligents.
La motivation premi\u00e8re de notre travail est de contribuer \u00e0 la r\u00e9alisation d’une adaptation plus centr\u00e9e sur l’humain dans les environnements intelligents. Nous d\u00e9finissons une telle adaptation comme une approche pour concevoir un syst\u00e8me intelligent dont l’adaptabilit\u00e9 consid\u00e8re des aspects de la subjectivit\u00e9 humaine.
Le domaine des Syst\u00e8mes Multi-Agents (SMA) est une approche prometteuse pour relever les d\u00e9fis de l’adaptabilit\u00e9 des syst\u00e8mes, tels que l’h\u00e9t\u00e9rog\u00e9n\u00e9it\u00e9 et la d\u00e9centralisation, le besoin de collaboration entre les services, et la capacit\u00e9 de planifier. Dans la litt\u00e9rature, les architectures d’agents courantes ont des raisonnements centr\u00e9s sur des aspects techniques, tels que les buts, plans et actions, ne pr\u00e9sentant pas les aspects de subjectivit\u00e9 humaine.
En explorant ces limitations, nous identifions la Th\u00e9orie des Valeurs Humaines Fondamentales. Elle a le potentiel d’am\u00e9liorer l’adaptation des mod\u00e8les de raisonnement actuels, \u00e9tant donn\u00e9 qu’elle se base sur les principes directeurs des personnes (les valeurs humaines). Par cons\u00e9quent, notre recherche se concentre sur la proposition d’une nouvelle approche, le SMASH, qui mod\u00e9lise un composant de raisonnement, une architecture d’agent, et un MAS. Une telle application a le potentiel d’atteindre un nouveau niveau d’adaptation collaborative centr\u00e9e sur l’humain dans les environnements intelligents.
<\/strong><\/p>\n

Abstract :<\/strong>Social spaces equipped with intelligent IoT (Internet of Things) devices are more common. These smart environments are made possible thanks to AI (Artificial Intelligence), and their goal is to make the lives of inhabitants more comfortable, improving the quality of life. Therefore, humans are a central aspect of smart environments.
The prime motivation of our work is to help bringing into fruition a more human-centered adaptation on smart environments. We define such adaptation as an approach to engineer an intelligent system, whose adaptability takes into account aspects of human subjectivity.
The AI domain of Multi-Agent Systems (MAS) is a promising approach to tackle the challenges in the adaptiveness of systems, such as the heterogeneity and decentralization in IoT, the need for collaboration among services, and the ability to plan for new situations. In the literature, common agent architectures have reasonings centered on technical aspects, such as goals, plans and actions, not presenting human subjectivity aspects as first-class citizens.
Exploring such limitations, we identified the Theory of Basic Human Values. It has the potential to enhance the adaptation on current reasoning models, since it is based on the guiding principles of people (the human values), related to their motivations and expectations. Therefore, our research focus on the proposition of a novel approach, the SMASH, in which we model a human values reasoning component, an agent architecture that uses such component, and a MAS incorporating this agent architecture. Such application has the potential to achieve a new level of collaborative human-centered adaptation in smart environments.<\/p>\n

Mots cl\u00e9s :<\/strong> agents intelligents, adaptation, environnements intelligents, intelligence artificielle, internet des objets, maison intelligente, syst\u00e8mes multi-agents, valeurs humaines.<\/p>\n

Keywords<\/strong> : adaptation, artificial intelligence, human values, intelligent agents, internet of things, multi-agent systems, smart environments, smart home.<\/p>\n

Date de Soutenance pr\u00e9vue : <\/strong> dernier trimestre de 2021<\/p>\n

Encadrement :<\/strong><\/p>\n