The proceedings of the 2nd edition of ALTRUIST have been published and are available at the link below:
CEUR-WS Proceedings: https://ceur-ws.org/Vol-3323
The workshop is scheduled for Friday 16th, December 2022, from 9:00 to 17:00 CET
|09:00 – 09:10||Opening||Francesca Fracasso|
|09:10 – 09:55||Designing an Emotion-Sensitive Companion Robot for the Elderly||Oliver Korn||Francesca Fracasso|
|09:55 – 10:10||Please ASTRO, can you follow me? Design of a social assistive robot for monitoring gait parameters||Alessandra Sorrentino, Niccolò Vezzi, Carlo La Viola, Erika Rovini, Filippo Cavallo, Laura Fiorini||Francesca Fracasso|
|10:10 – 10:25||An Environment to Collect Personal Memories of Older Adults and User them to Personalise Serious Games with Humanoid Robots||Benedetta Catricalà, Miriam Ledda, Marco Manca, Fabio Paterno, Carmen Santoro, Eleonora Zedda||Francesca Fracasso|
|10:30 – 11:15||Coffee Break|
|11:15 – 11:30||Design Thinking for a Robotic Poker Dealer||Madison R. Shippy, Abigail I.R. Leader, Naomi T. Fitter||Alessandro Umbrico|
|11:30 – 11:45||Identification of the highest wrinkle grasping point of a folded hospital gown||Olivia Nocentini, Jaeseok Kim, Julia Borras, Guillem Alenya, Filippo Cavallo||Alessandro Umbrico|
|11:45 – 12:00||Towards Adaptation of Humanoid Robot Behavior in Serious Game Scenarios using Reinforcement Learning||Eleonora Zedda, Marco Manca, Fabio Paterno||Alessandro Umbrico|
|12:00 – 12:15||Gerontosuit||Ivana Kowalikova||Alessandro Umbrico|
|12:15 – 12:30||Supporting behavior Editing for Social Robots||Riccardo De Benedictis, Gloria Beraldo, Gabriella Cortellessa, Francesca Fracasso, Amedeo Cesta||Laura Fiorini|
|12:30 – 12:45||Pedestrian and autonomous vehicle interaction: towards affective crossing||Domenico Sorrenti, Francesca Gasperini, Fabio D’Elia, Ballini Andrea, Simone Fontana, Federica Di Lauro, Alessandra Grossi, Stefano Dessena||Laura Fiorini|
|12:45 – 13:00||Towards Enhancing Social Navigation through Contextual and Human-related Knowledge||Phani Teja Singamaneni, Alessandro Umbrico, Andrea Orlandini, Rachid Alami||Laura Fiorini|
|13:00 – 14:00||Lunch Break|
|14:00 – 14:45||Conversational AI meets Social Robots. Towards Virtual Coaches for Wellbeing and Smart Aging||Kristiina Jokinen||Rainer Wieching|
|14:45 – 15:00||SI-ROBOTICS System: a preliminary study on usability of a rehabilitation program in patients with Parkinson’s disease||Roberta Bevilacqua, Marco Benadduci, Giovanni renato Riccardi, Giovanni Melone, Angela La Forgia, Nicola Macchiarulo, Luca Rossetti, Mauro Marzorati, Giovanna Rizzo, Pierpaolo Di Bitonto, Ada Potenza, Laura Fiorini, Federica Gabriella Cornacchia Loizzo, Carlo La Viola, Filippo Cavallo, Alessandro Leone, Gabriele Rescio, Andrea Caroppo, Andrea Manni, Amedeo Cesta, Gabriella Cortellessa, Francesca Fracasso, Andrea Orlandini, Alessandro Umbrico, Giulio Amabili, Lorena Rossi, Elvira Maranesi||Rainer Wieching|
|15:00 – 15:15||Investigating the role of different social cues in the human perception of a social robotic arm||Carlo La Viola, Laura Fiorini, Gianmaria Mancioppi, Filippo Cavallo||Rainer Wieching|
|15:15 – 15:30||Physiosmart: a preliminary study about the quality of rehabilitation using a computer vision approach||Fabio Tedone, Pierpaolo Di Bitonto, Davide Cafiero||Rainer Wieching|
|15:30 – 16:15||Social Applications of Multimodal Cognitive Robots||Alessandro Di Nuovo||Roberta Bevilacqua|
|16:15 – 16:20||Closing||Roberta Bevilacqua|
Please ASTRO, can you follow me? Design of a social assistive robot for monitoring gait parameters
Alessandra Sorrentino, Niccolò Vezzi, Carlo La Viola, Erika Rovini, Filippo Cavallo and Laura Fiorini
This paper proposes an alternative strategy for the analysis of the walking task using a socially assistive robot. This solution allows the person to avoid wearing additional sensors and allows the rehabilitation activity to be carried out both in facilities hospitals and in patients’ homes. In this work, we implemented a follow-me module to enable the robot to detect, track, and follow the patient during walking, adapting to his/her speed. The robot can recognize the person to be followed through a 2D laser sensor and an RGB-D camera: data from the laser are processed through the ROS package Leg Tracker, which extracts the 3-D position of the legs of the person; the data from the camera are used by the algorithm YOLOv3, which, by extracting the person’s bounding box, provides in output the 3D coordinates of the target to be tracked. In order for the robot to follow the patient at a predetermined distance, the follow-me module integrates PD and P controllers for handling the linear and angular velocities, respectively. The controllers’ gains were set according to the maximum speed attainable by the robot. The extracted walk parameters were compared with the parameters extracted by an inertial sensor placed on the feet (SensFoot) and analyzed to characterize the best robot configuration for the task of the gait assessment. Eleven participants were recruited to perform the tests with 3 different values of the robot’s maximum speed. For each test, 4 parameters were extracted from the laser and 10 parameters from the wearable sensors. The best configuration was found to be the one with the highest maximum speed, 0.7 m/s, whose gains from the two linear and angular controllers are Kp = 1.0, Kd = 0.4 and Kp = 1.0, respectively. Qualitative results collected at the end of the test also confirm the 0.7 m/s as the optimal perceived velocity.
An Environment to Collect Personal Memories of Older Adults and Use them to Personalise Serious Games with Humanoid Robots
Benedetta Catricalà, Miriam Ledda, Marco Manca, Fabio Paterno, Carmen Santoro and Eleonora Zedda
One of the goals of Ambient Assisted Living (AAL) solutions is to be able to stimulate the cognitive resources of older adults. An innovative way to address such stimulation is the use of serious games delivered through humanoid robots, as they can provide an engaging way to perform exercises useful for training human memory, attention, processing, and planning activities. This paper presents an approach to supporting cognitive stimulation based on personal memories. The humanoid robot can exhibit different behaviours using various modalities, and propose the games in a way personalised to specific individuals’ requirements, preferences, abilities, and motivations, which can vary among older adults, and even dynamically evolve over time for the same person depending on changing user needs and health conditions. Using personal memories associated with facts and events that occurred in older adults life in the serious games can increase their engagement, and thus potentially reduce the cognitive training drop-out.
Design Thinking for a Robotic Poker Dealer
Madison R. Shippy, Abigail I. R. Leader and Naomi T. Fitter
Social interaction through games is an integral part of the human experience, but in the wake of the COVID-19 pandemic, gameplay is more likely to be restricted. To support easier gameplay facilitation in a variety of settings, we propose a novel robot that deals poker and encourages social interaction among players. This robot was created using the design thinking approach. Throughout this design process, we worked with a convenience population of card game players and one expert in card-dealing and comedy. We found that the population warmly received the design overall, but had reservations mostly related to competence of and trust in the system. This work contributes to the design of future robots for social play.
Identification of the highest wrinkle grasping point of a folded hospital gown
Olivia Nocentini, Jaeseok Kim, Julia Borras, Guillem Alenya and Filippo Cavallo
There are already more than one billion people over the age of 60, and the World Health Organization predicts that number will increase to 1.4 billion by the year 2030. As a result, the need for caretakers is increasing, which could make society in the future unable to provide it. In this scenario, the need for automated assistance increases as the global population ages. One area of robotics where robots have demonstrated tremendous promise in closely collaborating with people is service robotics. Hospitals, residences, and facilities for the elderly will all require the deployment of intelligent robotic agents to carry out regular tasks. Cloth manipulation is one such daily activity and represents a challenging area for a robot. The research goal of this paper focused on finding the grasping points of the highest wrinkle (from a later point of view) of a folded hospital gown to then unfold it and help dressing a patient. The wrinkle is detected using the Generative Grasping Convolutional Neural Network (GG-CNN2), while the approach to the cloth by a manipulator is obtained by designing a visual servoing algorithm that considers the input of the GG-CNN2. In conclusion, the results described in this paper tend to study by deep some AI-based approaches for cloth manipulation capabilities; in particular, we concentrated on studying how to identify the first wrinkle of a cloth by combining the visual servoing approach with a neural network.
Towards Adaptation of Humanoid Robot Behavior in Serious Game Scenarios using Reinforcement Learning
Eleonora Zedda, Marco Manca and Fabio Paterno
Repetitive cognitive training can be seen as tedious by older adults and cause participants to drop out. Humanoid robots can be exploited to reduce boredom and the cognitive burden in playing serious games as part of cognitive training. In this paper, an adaptive technique to select the best actions for a robot is proposed to maintain the attention level of elderly users during a serious game. The goal is to create a strategy to adapt the robot’s behaviour to stimulate the user to remain attentive through reinforcement learning. Specifically, a learning algorithm (QL) has been applied to obtain the best adaptation strategy for the selection of the robot’s actions. The robot’s actions consist of a combination of verbal and nonverbal interaction aspects. We have applied this approach to the behaviour of a Pepper robot for which two possible personalities have been defined. Each personality is exhibited by performing specific actions in the various modalities supported. Simulation results indicate learning convergence and seem promising to validate the effectiveness of the obtained strategy. Preliminary test results with three participants suggest that the adaption in the robot is perceived.
Supporting behavior Editing for Social Robots
Riccardo De Benedictis, Gloria Beraldo, Gabriella Cortellessa, Francesca Fracasso, Amedeo Cesta
The growing deployment of social robots requires the ability to adapt to the dynamic changes occurring in the real environments. These reactive behaviors, however, are often incapable of reasoning and predicting the effects of their actions in the next future. Therefore, they must be accompanied by forms of deliberative semantic/causal reasoning. The combination of the reactive and deliberative forms of reasoning, which resembles the dual process theory, raises the problem of entrusting tasks to the corresponding modules. Just as happens in biological systems, the tendency to assign activities, as much as possible, towards the lower abstraction layers, equips the systems with more responsive capabilities at the cost of making the reactive layers more difficult to implement. In this document, we will introduce an architecture that, inspired by the classic three-tier architecture, combines slow and fast forms of reasoning, allowing social robots to achieve complex and dynamic behaviors. Since entrusting tasks to the more reactive components complicates their implementation (e.g., it requires the definition of formal rules which may not adequately generalize to unforeseen scenarios), we aim to reduce the technicalities and, consequently, to facilitate to the developers the implementation of the reactive behaviors. By relying on recent achievements in natural language translation, we will describe our recent efforts to adopt Transformer-based architectures, allowing the replacement of formal rules with easier to write “stories”, defined through sequences of perceived events and actions, entrusting the system with the task of learning behaviors by generalizing from them.
Pedestrian and autonomous vehicle interaction: towards affective crossing
Domenico Sorrenti, Francesca Gasperini, Fabio D’Elia, Ballini Andrea, Simone Fontana, Federica Di Lauro, Alessandra Grossi, Stefano Dessena
In near future scenarios, self-driving vehicles will circulate in urban environments, and their behaviour should be adapted with respect to different types of pedestrians. In particular, vehicles should be able to provide effective feedback, especially when dealing with the most vulnerable people, such as older adults and impaired subjects. Within this perspective, this paper illustrates the experimental settings and protocol to study pedestrian and autonomous vehicle interaction, especially focusing on the safeness felt by each subject in different crossing conditions. To this end, besides traditional self assessment questionnaires and video recordings, movement and physiological data are collected as indicators of stress. From the analysis of this multimodal data, different classes of pedestrians could be defined, that will guide the definition of proper vehicle behaviour depending on their level of confidence and safety feeling. A preliminary data collection have been performed and is here described in a controlled urban-like crossing environment. Subjects of various ages were considered, as well as different dynamic behaviours of a properly prepared vehicle, running in both human-controlled and self-driving modes.
Towards Enhancing Social Navigation through Contextual and Human-related Knowledge
Phani Teja Singamaneni, Alessandro Umbrico, Andrea Orlandini, Rachid Alami
Robots acting in real-world environments usually need to directly or indirectly interact with humans. Interactions may occur at different levels of abstraction (e.g., process, task, physical), leading to different research challenges (e.g., task allocation, human-robot joint actions, robot navigation). In the context of social navigation, we propose a conceptual integration of task and motion planning to contextualize robot behaviors. The idea is to leverage the contextual knowledge of a task planner to dynamically contextualize navigation skills according to the humans a robot is supposed to interact with, and the tasks to be performed. Specifically, we propose a holistic model of tasks and humans and map task-level knowledge to motion-level knowledge in order to constrain the implementation of robot trajectories.
SI-ROBOTICS System: a preliminary study on usability of a rehabilitation program in patients with Parkinson’s disease
Roberta Bevilacqua, Marco Benadduci, Giovanni renato Riccardi, Giovanni Melone, Angela La Forgia, Nicola Macchiarulo, Luca Rossetti, Mauro Marzorati, Giovanna Rizzo, Pierpaolo Di Bitonto, Ada Potenza, Laura Fiorini, Federica Gabriella Cornacchia Loizzo, Carlo La Viola, Filippo Cavallo, Alessandro Leone, Gabriele Rescio, Andrea Caroppo, Andrea Manni, Amedeo Cesta, Gabriella Cortellessa, Francesca Fracasso, Andrea Orlandini, Alessandro Umbrico, Giulio Amabili, Lorena Rossi, Elvira Maranesi
In this paper, SI-ROBOTICS, a rehabilitation programme for people with Parkinson’s disease, is presented along with preliminary results. The SI-ROBOTICS system, consisting of a robotic platform, a game, wearable and environmental sensors, and an artificial intelligence algorithm, aims to support the treatment of Parkinson’s patients following a rehabilitation programme based on Irish dance practice. Nine patients were recruited in the study and underwent 16 sessions of the programme. The primary objective of the study was to evaluate the usability of the system. Secondly, the benefits in terms of improved walking, balance and reduced risk of falls were evaluated. Preliminary results suggest that the system has a good chance of success, as it was found to be usable and effective in treating conditions typical of Parkinson’s disease.
Investigating the role of different social cues in the human perception of a social robotic arm
Carlo La Viola, Laura Fiorini, Gianmaria Mancioppi, Filippo Cavallo
The use of robotic arms in the rehabilitation context is increasing, even though the social aspect of the interaction is often neglected. This work aims at filling this gap, introducing social capabilities in a scenario of human-robot interaction for rehabilitation. This work will define the best social configuration for a robotic arm, in terms of social cues. In particular, sound and eyes cues will be linked to social movements designed using Laban Movement Analysis. Various combinations of movements and social cues were combined and embedded on a robotic arm. Then 15 participants were recruited and asked to perform a human-robot interaction task (handover) with the robotic arm. A questionnaire was used to evaluate the user impressions in terms of Valence, Arousal, the Godspeed components of Animacy and Safety, and two custom questions related to movement fluidity and likeability. The results show that the best combination for this exercise is the one where both sound and eyes are present, even though the data show high scores of Arousal and Animacy for the configuration composed of only sound and movement.
Physiosmart: a preliminary study about the quality of rehabilitation using a computer vision approach
Fabio Tedone, Pierpaolo Di Bitonto, Davide Cafiero
Smartphone-based computer vision tools have shown potential for practical application in the field of telerehabilitation. To provide greater accessibility, there is a need to reduce the need for sensors while ensuring the accuracy of patient monitoring without compromising the user experience. In this paper, we propose a new validation of a smartphone-based pose detection tool during the performance of a rehabilitation exercise in which certain metrics (elbow angle) need to be calculated.