Estudos sobre robôs de serviço como tecnologia assistiva
Carregando...
Arquivos
Data
2022-07-04
Autores
Miura, Lucas Heiji de Moraes
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A população idosa brasileira apresenta crescimento acelerado, em comparação com outros grupos etários, e irá ultrapassar a população de crianças e adolescentes na próxima década. Esta mudança na estrutura etária do país evidência a prioridade de alinhar as necessidades atuais e futuras desta parcela da população. Novos campos tecnológicos vêm ganhando destaque e notoriedade, como a tecnologia assistiva, a qual busca auxiliar idosos e portadores de deficiência, propiciando-lhes maior independência, dignidade e segurança para realizar atividades de vida diária como se medicar, praticar exercícios, realizar ligações e atividades mais elaboradas, por meio da robótica assistiva. O presente estudo teve como finalidade realizar estudos de modelos de interação em robôs de serviços para a área assistiva, incluindo idosos e pessoas com necessidades especiais, e propor um modelo de interação considerando vantagens e limitações dos modelos estudados na literatura acadêmica. Foram selecionados e analisados 60 estudos de modo a compreender aspectos e características relevantes para a interação humano-robô (HRI). O desenvolvimento do modelo de interação humano-robô utilizou como base os artigos estudados e aplicou oportunidades de melhoria ao integrar um robô físico a um assistente virtual, através do uso da ferramenta AWS RoboMaker do Alexa Skill Set (ASK). Robôs de serviço podem ter diferentes fisionomias e funções, assim como podem desempenhar de atividades simples até às mais complexas. O design do robô influencia nas expectativas atribuídas a ele e afeta diretamente a interação humano-robô. Idosos demonstram preferências por robôs com distinção clara de nível hierárquico, com 68,4% dos sistemas apresentando robôs subordinados, como máquinas, servos e assistentes; por robôs de aparência antropomórfica, que representam 55% dos robôs avaliados de projetos cujos idosos equivalem a 60% do público-alvo, principalmente se disporem de detalhes minuciosos de expressão social, habilidade presente em 70,5% dos robôs analisados, os quais impactam de forma positiva a interação. A necessidade dos graus de liberdade está diretamente relacionada com a função que o robô deve desempenhar. Robôs humanoides apresentam maior quantidade de graus de liberdade, com os diferentes robôs humanoides avaliados apresentando uma média de 14 DoF, uma vez que buscam replicar características e ações humanas complexas e de forma fluente.
The Brazilian elderly population is growing rapidly, compared to other age groups, and will soon surpass the population of kids and adolescents in the next decade. This shift in the country’s age structure highlights the priority to align the current and future needs of this part of the population. New technological fields have been gaining prominence and notoriety, such as assistive technology, which seeks to help the elderly and people with disabilities, providing them greater independence, dignity and safety to perform activities of daily living such as taking medication, exercising, making phone or video calls and more elaborate activities, through assistive robotics. This study aimed to carry out studies of interaction models in service robots for the assistive area, including the elderly and people with special needs, and to propose an interaction model considering advantages and limitations of the models studied in the academic literature. Sixty studies were selected and analyzed to understand aspects and characteristics relevant to human-robot interaction (HRI). The development of the human-robot interface model was based on the articles studied, and applied opportunities for improvement by integrating a physical robot with a virtual assistant, using the AWS RoboMaker tool from the Alexa Skill Set (ASK). Service robots can have different physiognomies and functions, as they can perform from simple to more complex tasks. Robot design influences the expectations assigned to it and directly affects HRI. It’s not enough for the robot to be efficient, it must convey a sense of confidence to the user and be robust to operate autonomously in case something happens to the user. SLAM was popularly used for navigation systems; the LBPH technique showed popularity for facial recognition tasks; as well as the Haar Cascade algorithm for object detection. Elderly people show preferences for robots with a clear distinction of hierarchical level, with 68.4% of systems featuring subordinate robots, such as machines, servants and assistants; and by robots with an anthropomorphic appearance, which represent 55% of the evaluated robots of projects whose elderly people are equivalent to 60% of the target audience, especially if they display minute details of social expression, a skill present in 70.5% of the analyzed robots, which positively impact the interaction. The need for degrees of freedom is directly related to the role that the robot must perform. Humanoid robots have a greater number of degrees of freedom, with the different humanoid robots evaluated presenting an average of 14 DoF, as they seek to replicate complex human characteristics and actions fluently.
The Brazilian elderly population is growing rapidly, compared to other age groups, and will soon surpass the population of kids and adolescents in the next decade. This shift in the country’s age structure highlights the priority to align the current and future needs of this part of the population. New technological fields have been gaining prominence and notoriety, such as assistive technology, which seeks to help the elderly and people with disabilities, providing them greater independence, dignity and safety to perform activities of daily living such as taking medication, exercising, making phone or video calls and more elaborate activities, through assistive robotics. This study aimed to carry out studies of interaction models in service robots for the assistive area, including the elderly and people with special needs, and to propose an interaction model considering advantages and limitations of the models studied in the academic literature. Sixty studies were selected and analyzed to understand aspects and characteristics relevant to human-robot interaction (HRI). The development of the human-robot interface model was based on the articles studied, and applied opportunities for improvement by integrating a physical robot with a virtual assistant, using the AWS RoboMaker tool from the Alexa Skill Set (ASK). Service robots can have different physiognomies and functions, as they can perform from simple to more complex tasks. Robot design influences the expectations assigned to it and directly affects HRI. It’s not enough for the robot to be efficient, it must convey a sense of confidence to the user and be robust to operate autonomously in case something happens to the user. SLAM was popularly used for navigation systems; the LBPH technique showed popularity for facial recognition tasks; as well as the Haar Cascade algorithm for object detection. Elderly people show preferences for robots with a clear distinction of hierarchical level, with 68.4% of systems featuring subordinate robots, such as machines, servants and assistants; and by robots with an anthropomorphic appearance, which represent 55% of the evaluated robots of projects whose elderly people are equivalent to 60% of the target audience, especially if they display minute details of social expression, a skill present in 70.5% of the analyzed robots, which positively impact the interaction. The need for degrees of freedom is directly related to the role that the robot must perform. Humanoid robots have a greater number of degrees of freedom, with the different humanoid robots evaluated presenting an average of 14 DoF, as they seek to replicate complex human characteristics and actions fluently.
Descrição
Palavras-chave
Interação homem-maquina, Tecnologia assistiva, Idosos, Robôs, Assistive technology, Elderly