2022 – Present
People with dementia (PWD) have low involvement in social/leisure activities. We aim to evaluate the effects of social robots on the engagement of PWD residing in long-term care facilities. Thirty institutionalized PWD (confirmed by the Mini-Mental State Examination) will be randomized into two groups: usual activities and robot interaction. A cognitive engagement scale will be applied before, during, and after the interaction. Results may enhance the use of robots in the care of PWD.
2022 – Present
With the increase in the elderly population and life expectancy, healthy aging is a new challenge. However, aging has often been understood as a burden on society, involving tasks related to the demand for pensions, healthcare, and long-term care services. The COVID-19 pandemic had a disproportionate impact on residents of long-term care facilities (LTCFs) and exposed persistent inequalities. The WHO encouraged the maintenance of social connections through digital alternatives, but it became evident that there is little support available to assist those who need help using technology for these purposes. We hypothesize that initiatives aimed at technological literacy, supported by the field of gerontechnology, using social robots, may be a promising path to access new learning opportunities and promote digital literacy for elderly residents in LTCFs. Forty residents aged 60 years or older, without cognitive deficit or functional impairment, will participate. They will be randomized into a control group, which will carry out the routine activities of the LTCF, and an experimental group, which will interact with the social robot twice a week for 50 minutes each time, over 12 weeks.
Instruments:
Procedures:
For 50 minutes, interactive activities with the robot will be proposed. The activities to be programmed will be categorized by current events, entertainment, culture and art, and social interaction. For the control group, a routine LTCF activity will be used at least twice a week, so that participants allocated to this group will also be filmed before, during, and after the activity, following the same procedures as the experimental group.
2022 – Present
Smart connected toys are increasingly present in children’s lives, representing a significant market niche for toy manufacturers. Advances in human-computer interaction and Artificial Intelligence (AI) have already been incorporated into smart toys to offer greater autonomy and inductive reasoning skills via machine learning. However, the possibility of smart toys being biased due to machine learning has been overlooked. Machine learning typically relies on a dataset of examples to extract models that represent the real world; however, if these data are biased, the resulting behavior of the machine learning-based system will naturally contain biases. With prejudice embedded in the dataset, the toy learns this prejudice and may act with discrimination. Although bias is a general problem in machine learning, it becomes more serious when the users are children, who are much more sensitive to biased patterns. This research project aims to deepen our understanding of the relationship between smart toys and biased machine learning to propose sociotechnical solutions for designing a broad set of artifacts that can be adopted by smart toy manufacturers to support the development of unbiased toys with a low risk of motivating discriminatory behavior in the end user. We hope to contribute to preventing the toy industry from producing biased smart toys, as well as to advancing the field related to machine learning enabled fairness.
2021 – Present
The project involves teaching programming to teenagers with a focus on understanding concepts and applications of artificial intelligence and companion robots. The objectives are to create educational kits for teaching programming logic based on the concept of programming challenges. These challenges include: A problem that must be solved through programming; A motivating context, usually based on real-life situations, that frames the problem to be solved; select AI concepts and applications that can be introduced to students within the motivating context mentioned above
and to develop didactic strategies that explain this content without requiring prerequisites for using the material. The idea is to follow the principles of human-centered artificial intelligence, bringing in concepts that minimize the harmful effects of new technologies and maximize positive outcomes. The same approach applies to concepts and applications related to companion robots.
Regarding the use of companion robots in the project, the project proposer possesses two companion robots (Zenbo and Zenbo Jr. by ASUS) used in scientific research. These robots will be used in the project to illustrate concepts, including teaching programming through a block programming environment (open access) for these robots. The material created will be tested and evaluated with students from municipal schools in the East Zone of São Paulo, either in person or online.
2020 – Present
Socially Assistive Robots (SARs) have been introduced to support individuals with specific needs in society. Service robots, for example, can assist elderly people in their homes with tasks ranging from routine activities like medication reminders to more demanding tasks such as assessing depression levels for basic screening and subsequently alerting family members and medical specialists. Companion robots can also aid children with autism spectrum disorder in developing interaction skills with others. These social robot activities depend on the degree of intelligent functions they can deliver, which has significantly improved in recent years. Advances in robotics technology and the evolution of artificial intelligence present great potential for providing social robots with decision-making and cognitive capabilities.
Social robots are not yet as prevalent in Brazilian society, partly due to factors such as distance from technological development centers and lower economic and cultural development among the civilian population. Despite this, the integration of social robots in human interactions, including the potential replacement of certain types of labor, seems inevitable. However, the lack of experience across different sectors of Brazilian society regarding the positive or negative effects of this technology could lead to economic losses for the country. Amid the current global health crisis, robotic technology is increasingly seen as essential, including in facilitating human interactions.
This project aims to serve as a catalyst for establishing public policies to promote and regulate this type of technology, as well as supporting the development of Brazilian technology for robot use.
2019 – 2019
Several recent medical reports reveal that depression and anxiety have a serious impact on health (e.g., increased death rate of older people, increased risk of stroke etc.). Thus, it is important for societies to proactively address depression, especially in independently living seniors. To this end, “robots” are increasingly taking their places around us to help to address many challenges issues including companionship and even in-home healthcare. This project aims to address the elderly individuals with depression and anxiety in Brazil by developing means in mindfulness meditation for alleviating the in-home caretakers from routine tasks that can be partially automated and thus offloaded to companion robots. This project will methodically examine Brazilian cultural and ethical content with regard to perceptions of mindfulness meditation through companion robots for elderly individuals with depression. The key research outcomes will include a fully developed environment for Android-based robots and recommendations for the robot industry.
2016 – 2016
Children’s toys have become increasingly sophisticated over the years, with a growing shift from simple physical products to toys that engage the digital world by the use of software and hardware. A smart anthropomorphic toy is defined as a device consisting of a physical toy component in a humanoid form that connects to a computing system with online services through networking and sensory technologies to enhance the functionality of a traditional toy. The main objective of this research is on developing a privacy-aware context data model for smart toys to support a standardized child protection framework with parental controls. The framework includes an alert mechanism by applying text mining techniques to identify suspicious dialogues between children and the toys. The potential impact of this research is to provide a safe smart toy computing delivery model to protect children.
2014 – 2016
A security and privacy model for a mobile computing application is an important and challenging topic in the area of service computing research. In the proposed research project, we focus on the security and privacy enforcement model for m-services in toy computing from the perspective of policy framework. A toy is a product that is intended for use by a child in learning or play. The toy industry is comprised of establishments primarily engaged in manufacturing dolls, toys and games. Toy companies are confronted with the challenge of better understanding the consumer needs, concerns and exploring the possibility of adopting such context-aware wearable toys to information interfaces that will successfully market toy computing. An exacerbating issue is that a privacy policy framework with technological standards for protecting players’ location based information on m-services has not yet been established within the toy industry worldwide. Thus, m-services need to be highly customized to tackle security and privacy preferences in order to support toy computing.
