Publication
Assistência à autonomia no domicílio com integração de automação e inteligência artificial
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.99 MB | Adobe PDF |
Advisor(s)
Abstract(s)
O Ambient Assisted Living (AAL) cresce em importância na investigação e
investimentos públicos e privados, principalmente pelo envelhecimento da
população mundial. Neste contexto, um problema que se busca resolver é
a identificação e representação de atividades do cotidiano e para isto a Representação
do Conhecimento mostra-se útil. A oferta de novas tecnologias
Internet of Things (IoT) permitiu a criação de diversos protocolos de comunicação
sem o suporte de um padrão, dificultando a integração de dispositivos
em um único ambiente. Uma forma de solucionar este problema se dá na
forma de aplicações de automação que traduzem estes protocolos em uma
abstração que permita a interação com os dispositivos de modo uniforme.
Esta disseminação de dispositivos IoT também disponibilizou uma quantidade
massiva de dados para serem tratados, justificando o interesse de técnicas
de Aprendizagem Máquina.
Este trabalho especifica, implementa e valida um sistema de ambiente inteligente
que agrega automação e Inteligência Artificial (IA). A especificação
dá-se na forma de uma arquitetura de referência, composta por três módulos,
cujas tarefas são automatizar e uniformizar a recolha de dados, relacionar e
dar significado a estes dados e aprender com estes. São propostos três casos
de uso, que representam as tarefas desempenhadas por cada um dos módulos,
tendo sido formulados de modo a demonstrar uma gradual integração destes
componentes do sistema. A implementação trata das aplicações e técnicas
utilizadas para o desenvolvimento de um protótipo funcional deste sistema.
A validação é realizada com a avaliação dos testes realizados para cada um
dos casos de uso propostos.
É apresentada uma discussão sobre temas relacionados ao trabalho, especificamente
sobre aplicações de automação, Representação do Conhecimento,
suas diferentes linguagens, Aprendizagem Máquina e são analisados trabalhos
e artigos destas áreas. O resultado do trabalho é a especificação de uma
arquitetura de referência para um sistema de acompanhamento de idosos que
moram sozinhos e apresentados três casos de uso que representam as tarefas
de cada um dos módulos.
Ambient Assisted Living (AAL) grows in importance in scientific researches and investment by companies and governments, mainly due to the aging of the world population. In this context, a problem that is sought to solve is the identification and representation of activities of daily living. For this purpose the Knowledge Representation proves itself useful. The offer of new technologies known as Internet of Things (IoT) result in the creation of several communication protocols, making difficult to integrate devices from different vendors into a single environment. One way to solve this problem is the use of a middleware layer that integrates these protocols into an abstraction and allows interaction with devices in a uniform manner. This dissemination of IoT devices also triggered a large quantity of data to be processed, justifying the use of Machine Learning (ML) techniques. This work specifies, implements, and validates a smart environment system that aggregates Automation and Artificial Intelligence (AI). The specification presents a reference architecture, composed by three modules, whose tasks are to automate and standardize the collection of data, to relate and give meaning to that data and to learn from it. Three use cases are proposed and represent the tasks performed by each of the modules, have been formulated in such a way as to demonstrate a gradual integration of these components of the system. The implementation addresses the applications and techniques used for the development of a functional prototype of this system. The validation is performed with the evaluation of the tests performed for each of the proposed use cases. A discussion on work-related topics is presented, specifically on automation applications, Knowledge Representation, their different languages, ML and are analyzed papers and articles of these areas. The result of the work is the specification of a reference architecture for a monitoring system for elderly people who live alone and are presented three use cases that represent the tasks of each of the modules.
Ambient Assisted Living (AAL) grows in importance in scientific researches and investment by companies and governments, mainly due to the aging of the world population. In this context, a problem that is sought to solve is the identification and representation of activities of daily living. For this purpose the Knowledge Representation proves itself useful. The offer of new technologies known as Internet of Things (IoT) result in the creation of several communication protocols, making difficult to integrate devices from different vendors into a single environment. One way to solve this problem is the use of a middleware layer that integrates these protocols into an abstraction and allows interaction with devices in a uniform manner. This dissemination of IoT devices also triggered a large quantity of data to be processed, justifying the use of Machine Learning (ML) techniques. This work specifies, implements, and validates a smart environment system that aggregates Automation and Artificial Intelligence (AI). The specification presents a reference architecture, composed by three modules, whose tasks are to automate and standardize the collection of data, to relate and give meaning to that data and to learn from it. Three use cases are proposed and represent the tasks performed by each of the modules, have been formulated in such a way as to demonstrate a gradual integration of these components of the system. The implementation addresses the applications and techniques used for the development of a functional prototype of this system. The validation is performed with the evaluation of the tests performed for each of the proposed use cases. A discussion on work-related topics is presented, specifically on automation applications, Knowledge Representation, their different languages, ML and are analyzed papers and articles of these areas. The result of the work is the specification of a reference architecture for a monitoring system for elderly people who live alone and are presented three use cases that represent the tasks of each of the modules.