Personal social robots are intelligent and socially interactive systems designed to support daily life through assistance, companionship, and personalized interaction. Central to this vision is the ability to tailor behavior, communication, and support to users’ needs, routines, and emotional states over time. This paper addresses the challenge of supporting personalization by modeling the long-term memory of a social robot as a multi-domain, ontology-grounded Knowledge Graph, which is automatically updated and managed through a Large Language Model during human-robot interaction. The proposed approach focuses on incremental knowledge acquisition, ontology-level consistency management, and coherent personalization across domains throughout the interaction. The framework was integrated into the UBTECH Alpha Mini robot and evaluated in a user study with thirty participants, comparing a static system in which the Knowledge Graph remains unchanged during interaction with a dynamic version employing LLM-based Knowledge Graph management. Results were analyzed from two complementary perspectives. From a system-level viewpoint, most Knowledge Graph completion and update operations resulted in correct and semantically consistent knowledge extensions. From a user-centered perspective, participants significantly perceived the dynamic robot as more adaptive and personalized, demonstrating the effectiveness of automated Knowledge Graph updates in supporting personalization and user adaptation in social robots.

Cognitive Stimulation Therapy (CST) is an evidence-based intervention that involves group or individual sessions demonstrating cognitive benefits in Mild Cognitive Impairment. Recent research suggests that social robots can effectively assist therapists during cognitive stimulation sessions for people with MCI. Building on these findings, we conducted an experimental study to evaluate the impact of a Socially Assistive Robot, on engagement and emotional responses during cognitive interventions for a group of elderly people. Each session was video-recorded for subsequent analysis. The findings suggest that the use of a Social Robot as a mediating tool in CST interventions is associated with high levels of participant engagement and predominantly positive-valenced emotional responses, as detected by both human and automatic evaluations.

This paper presents the development of CArEN (Conversational AgEnt supporting Narrative medicine) that integrates a text-based emotional recognition module to personalize therapeutic pathways in the context of Narrative-Based Medicine (NBM).NBM combines traditional medicine, therapies, symptom monitoring, and vital parameters detection with a conversation-based approach that allows considering not only physical well-being but also the psychosocial and emotional impact of illness on the patient's life.A study was carried out to evaluate the models' effectiveness in real-world contexts and collect user feedback on the conversational agent's performance and empathic support.The results demonstrate good accuracy in emotion recognition and positive user feedback, highlighting the conversational agent's potential as an effective means of supporting narrative medicine techniques.

Childhood obesity and overweight are concerning issues worldwide with significant health and social implications, making it necessary to implement prevention and intervention policies. Engaging children through creative and playful methods is an effective approach to motivating behavior change. Social robots have emerged as promising tools in education and persuasive technology, due to their interactive and engaging nature. This paper presents a study examining the potential of a social robot acting as a nutrition coach for children, by comparing interactions with a social robot versus a chatbot. Results suggest that, in the short term, the social robot was perceived as more persuasive, increasing participants’ intention to adopt healthier eating behavior. These findings offer insights for the development of personalized nutrition coaching for children.

Engaging in regular exercise offers numerous mental and physical health benefits. Many people, after the Covid-19 pandemic, started to work out at home. However, performing exercises incorrectly can result in injuries. For this reason, technology can be an effective tool for encouraging physical activity at home, and social robots have emerged as a potential solution for providing fitness training in a safe and engaging manner. A physical robot is often perceived as more socially attractive compared to virtual agents. Despite the growing popularity and existing literature on social robotics-based fitness solutions, there is a lack of research examining the accuracy of these systems and interventions at home during physical exercise recognition. In this paper, we propose a novel approach utilizing the Ubtech alpha mini social robot, which possesses motor capabilities to demonstrate exercises. We endowed the robot with the capability to recognize exercise correctness and motivate users to engage in physical activities. In this project, first of all, a new dataset of physical exercises was created, including 1,500 videos of physical exercises. Then, a new approach to recognize physical exercises from image sequences was presented. This method was implemented in a personal social robot to recognize physical exercises using a deep learning model. The results of the study validate the effectiveness of the system in recognizing physical exercises through the camera onboard a personal social robot.

The automatic monitoring and assessment of the engagement level of learners in distance education may help in understanding problems and providing personalized support during the learning process. This article presents a research aiming to investigate how student engagement level can be assessed from facial behavior and proposes a model based on Long Short-Term Memory (LSTM) networks to predict the level of engagement from facial action units, gaze, and head poses. The dataset used to learn the model is the one of the EmotiW 2019 challenge datasets. In order to test its performance in learning contexts, an experiment, involving students attending an online lecture, was performed. The aim of the study was to compare the self-evaluation of the engagement perceived by the students with the one assessed by the model. During the experiment we collected videos of students behavior and, at the end of each session, we asked students to answer a questionnaire for assessing their perceived engagement. Then, the collected videos were analyzed automatically with a software that implements the model and provides an interface for the visual analysis of the model outcome. Results show that, globally, engagement prediction from students' facial behavior was weakly correlated to their subjective answers. However, when considering only the emotional dimension of engagement, this correlation is stronger and the analysis of facial action units and head pose (facial movements) are positively correlated with it, while there is an inverse correlation with the gaze, meaning that the more the student's feels engaged the less are the gaze movements.

Abstract The attention towards robot-assisted therapies (RAT) had grown steadily in recent years particularly for patients with dementia. However, rehabilitation practice using humanoid robots for individuals with Mild Cognitive Impairment (MCI) is still a novel method for which the adherence mechanisms, indications and outcomes remain unclear. An effective computing represents a wide range of technological opportunities towards the employment of emotions to improve human-computer interaction. Therefore, the present study addresses the effectiveness of a system in automatically decode facial expression from video-recorded sessions of a robot-assisted memory training lasted two months involving twenty-one participants. We explored the robot’s potential to engage participants in the intervention and its effects on their emotional state. Our analysis revealed that the system is able to recognize facial expressions from robot-assisted group therapy sessions handling partially occluded faces. Results indicated reliable facial expressiveness recognition for the proposed software adding new evidence base to factors involved in Human-Robot Interaction (HRI). The use of a humanoid robot as a mediating tool appeared to promote the engagement of participants in the training program. Our findings showed positive emotional responses for females. Tasks affects differentially affective involvement. Further studies should investigate the training components and robot responsiveness.

The fall prevention in the elderly population is a field of growing interest. This paper presents the preliminary results of a fall prevention system based on a customized exergame program. Results show that the participants involved in the experiments evaluate positively the system usability. Moreover, in order to evaluate the efficiency of the system, a global improvement of around 8.8% has been observed in the postural response after just two sessions with the system.

This paper describes the first prototype of a gaze-based system designed for providing interactively information about dresses shown, on physical mannequins, in a shop window. Using the system the user may look at available sizes, colors, price and similar products. Due to the nature of such a system, the interaction must be touchless and natural. The developed solution uses Microsoft Kinect 2 as a device to achieve a natural gaze-based pointing approach. Results show that users evaluate positively the approach and felt positively engaged during the interaction.