Modelo sensível à situação para tomada de decisão multiobjetiva em ambientes inteligentes

• Ano de defesa: 2018
• Autor(a) principal: Teixeira, Milene Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Ciência da Computação; UFSM; Programa de Pós-Graduação em Informática; Centro de Tecnologia
• Palavras-chave em Português: Multiobjetividade; Sensibilidade a situação; Sensibilidade ao contexto; L-fuzzy; Ambientes inteligentes; Multiobjectivity; Situation awareness; Context awareness; Ambient intelligence; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: http://repositorio.ufsm.br/handle/1/15027

More and more there is the emergence of automated environments that seek to add intelligence to the decision making on the many different real world problems, which often have multiple and conflicting objectives. In some cases, these objectives have the same importance and, therefore, one or the other cannot be prioritized. For dealing with human reasoning and everyday activities, Ambient Intelligence (AmI) systems often present this feature. This paper presents a situation aware model that aims to assist systems in multiobjective decision making for objectives with same importance in AmI. From context data obtained from sensors in the environment, a system developed based on this model identifies the situation of interest; performs multiobjective decision making without assigning weights to the objectives; and automatically performs an action to control the environment. To verify the proposed model, a system was developed using as scenario an office in an AmI for which it is desired to provide thermal comfort to the user while avoiding unnecessary energy consumption. Allied to this system, it was used an IoT device that makes use of sensors and is able to: a) obtain data on consumption and temperature of the environment and b) manipulate the temperature setting of the air conditioner. As results, this work provides an L-fuzzy library (used in the decision module) and shows that the inclusion of this intelligence in AmI systems allows the achievement of both objectives, without the need to choose one or the other aspect.