Socially assistive robots (SARs) have shown great promise in supplementing and augmenting interventions to support the physical and mental well-being of older adults. However, past work has not yet explored the potential of applying SAR to lower the barriers of long-term low vision rehabilitation (LVR) interventions for older adults. In this work, we present a user-informed design process to validate the motivation and identify major design principles for developing SAR for long-term LVR. To evaluate user-perceived usefulness and acceptance of SAR in this novel domain, we performed a two-phase study through user surveys. First, a group (n = 38) of older adults with LV completed a mailed-in survey. Next, a new group (n = 13) of older adults with LV saw an in-clinic SAR demo and then completed the survey. The study participants reported that SARs would be useful, trustworthy, easy to use, and enjoyable while providing socio-emotional support to augment LVR interventions. The in-clinic demo group reported significantly more positive opinions of the SAR's capabilities than did the baseline survey group that used mailed-in forms without the SAR demo.
As a new type of human companion, social robots are becoming more and more popular and expected to being fully integrated with human daily life in the near future. Being able to correctly perceive the emotions of users and react to it can increase the sense of trust, affinity, and social presence of human-robot interaction. In this paper, we propose a human-centered reinforcement learning strategy to train social robots to achieve autonomous emotion understanding and behavior shaping. Our whole study was conducted on the social robot Haru, which has a large library of routines to express different emotions. Our experimental results show that autonomous emotion understanding and behavior shaping of social robots can be achieved through continuous interaction with humans.
Social robots are increasingly used in learning settings. So far, the main focus of this has been in school lessons and teaching at universities. Another possible setting is the children’s hospital. There, for example, young patients need to acquire basic knowledge about their disease so that they can deal with it appropriately. This should be done in a joyful, fun way, as the situation is stressful enough in itself, and so learning is also facilitated. The paper presents a learning application for diabetic children that runs on Pepper. This social robot was particularly well suited for this task because it has a large integrated touchscreen, similar to a tablet. A learning game is displayed on it that was developed especially for this setting. The children have to estimate the carbohydrate values of foods and meals or answer knowledge questions. The social robot gives verbal and gestural feedback in each case. The subjects responded overwhelmingly positively to the learning application. Pepper’s visible and audible feedback plays a special role in this. Social robots like Pepper are an interesting solution for knowledge transfer in a children’s hospital.
As integrating social robots in elderly care scenarios becomes increasingly prevalent, the need for ethical decision-making frameworks to govern their actions is critically important. This paper presents a comprehensive computational approach using supervised machine learning algorithms to address the ethical considerations inherent in robot-assisted fetching tasks for the elderly. Drawing upon established ethical principles and novel moral dimensions specific to elderly care, we develop an intricate framework encompassing diverse entities and scenarios using a greet or beat approach. To validate the framework, we conducted a pilot study involving thirty participants experienced in caregiving. Through an interactive application, participants designed scenarios, decided whether the robot should fetch objects, and provided reasons for their choices. Their decisions were then compared with predictions generated by a set of machine learning algorithms trained on a dataset of various scenarios. Our results shed light on the diverse ethical perspectives in elderly care and the feasibility of automating ethical decision-making for social robots in this domain. This research contributes to the burgeoning field of roboethics, offering insights and tools to guide the responsible deployment of robots in assistive elderly care, ultimately promoting the well-being and ethical treatment of elderly individuals.
As the use of robotized products becomes more prevalent in real-world settings, users are confronted with unforeseen problematic situations. The response strategy of a robotized product can have a significant impact on mitigating negative perceptions of the product resulting from problematic situations. In this study, we introduce apology and risk communication strategies as potential response strategies for robotized products and explore their impact on user evaluations or attraction through experimental research. Our findings indicate that when a problem originates from the robotized product itself, the use of an apology strategy and an accommodative risk communication strategy can enhance users’ task attraction to the product, but do not affect their evaluation of the robot and social attraction to it. Conversely, when the problem is not caused by the robotized product, the use of a defensive risk communication strategy is associated with higher robot evaluation and greater user attraction compared to the use of an accommodative strategy. This study demonstrates that response strategies in problematic situations that people expect for robotized products are different from those for humans.
Drones and other robotic technologies enable us to explore and study the world without the need for human guidance. This enables us to interact with previously uncharted areas that pose a risk to human safety, including the underwater biosphere. A drone may encounter a variety of problems when operating in such an unpredictable and hazardous environment. In these contexts, creating a digital twin of a marine drone can provide several advantages, such as for example to enhance the management, maintenance, and performance of the marine drone through data-driven approaches, benefiting efficiency and safety. In this work, we present the development of a digital twin prototype, which is a virtual representation of a physical entity of a marine drone-type unmanned surface vehicle. ROS, Blender, and the AWS cloud platform were used to create the system. The three-dimensional model of the maritime drone was created in Blender, and the AWS Robomaker service was utilized for simulation testing and possible deployment of the robotic application without managing any infrastructure. The goal is to present the tools and architecture that were developed to collect data using non-invasive ways and generate a marine digital twin.
The role of trust in human-robot interaction (HRI) is becoming increasingly important for effective collaboration. Insufficient trust may result in disuse, regardless of the robot’s capabilities, whereas excessive trust can lead to safety issues. While most studies of trust in HRI are based on questionnaires, in this work it is explored how participants’ trust levels can be recognized based on electroencephalogram (EEG) signals. A social scenario was developed where the participants played a guessing game with a robot. Data collection was carried out with subsequent statistical analysis and selection of features as input for different machine learning models. Based on the highest achieved accuracy of 72.64
This paper presents an exploration of the role of explanations provided by robots in enhancing transparency during human-robot interaction (HRI). We conducted a study with 85 participants to investigate the impact of different types and timings of explanations on transparency. In particular, we tested different conditions: (1) no explanations, (2) short explanations, (3) detailed explanations, (4) short explanations for unexpected robot actions, and (5) detailed explanations for unexpected robot actions. We used the Human-Robot Interaction Video Sequencing Task (HRIVST) metric to evaluate legibility and predictability. The preliminary results suggest that providing a short explanation is sufficient to improve transparency in HRI. The HRIVST score for short explanations is higher and very close to the score for detailed explanations of unexpected robot actions. This work contributes to the field by highlighting the importance of tailored explanations to enhance the mutual understanding between humans and robots.
Autonomous driving technologies can minimize accidents. Communication from an autonomous vehicle to a pedestrian with a feedback module will improve the pedestrians’ safety in autonomous driving. We compared several feedback module options in a Virtual Reality environment to identify which module best increases public acceptance, legibility, and trust in the autonomous vehicle’s decision, and to identify preference. The results of this study show that participants prefer symbols or text over lights and road projection with no significant difference between symbols and text. Further, our results show that the preferred text interaction mode option when the vehicle is not driving is “Walk,” “Safe to cross,” “Go ahead” and “Waiting”, and the preferred symbol interaction mode option is the walking person as on a traffic light, with no significant preference between the cross advisory symbol and the pedestrian crossing sign.
One of the major issues in pediatric rehabilitation practices relates to children refusing to participate in or perform associated exercises targeted to improve their physical condition. Technology and serious games are effective approaches to engage and motivate children and assist therapists in rehabilitation exercises. This Paper tries to elicit from children's requirements for the objective of designing efficient serious games scenarios that facilitate the rehabilitation procedure. A novel set of six rehabilitation game scenarios on standing frame for robotic assistance involving children with cerebral palsy is presented. We discuss the use of serious games on a standing frame in terms of humanoid robot limitations and capabilities. The scenarios have been developed based on specialists’ observations and in situ consultations with therapists at a pediatric rehabilitation center. Our findings are expected to help in future research tailored toward studying the effectiveness of adding humanoid robots to rehabilitation games to increase children’s motivation, engagement, and enjoyment.
Persuasive robotics has gained immense traction over the years, due to its potential to be used as a behavioral change system and to influence decision making in humans. With a plethora of existing studies in the field, this paper adds to the sea of knowledge by exploring the effect of number of robots on perceived persuasion and competence of a robot by revisiting Human-human Interaction theories of ‘multiple-source effect’ and ‘message reinforcement’. A simple two condition (one vs two robots) between-subjects experiment was conducted across two stages, and human participants engaged in a persuasive dialog-based interaction with the robot(s) Pepper and NAO where participants choose between two drink options and complete a survey. The results reveal a single robot is more persuasive and competent than multiple robots. Further analysis of qualitative data provides insights about the effect of robot morphology, social influence, familiarity with technology and intergroup dynamics, all of which collectively impact human perceptions and reactions. The implications of this research showcase the purposeful use of robots for marketing or brand promotion and further encourage better strategies of natural robot-robot & human-robot interactions.
This paper discusses pilot deployment of a social robot “WallBo” that investigated the effectiveness in promoting and encouraging handwashing practices among children in a rural school in India. The results suggest an overall 85.06
Given the importance of gaze in Human-Robot Interactions (HRI), many gaze control models have been developed. However, these models are mostly built for dyadic face-to-face interaction. Gaze control models for multiparty interaction are more scarce. We here propose and evaluate data-driven gaze control models for a robot game animator in a three-party interaction. More precisely, we used Long Short-Term Memory networks to predict gaze target and context-aware head movements given robot’s communication intents and observed activities of its human partners. After comparing objective performance of our data-driven model with a baseline and ground truth data, an online audiovisual perception study was conducted to compare the acceptability of these control models in comparison with low-anchor incongruent speech and gaze sequences driving the Furhat robot. The results show that our data-driven prediction of gaze targets is viable, but that third-party raters are not so sensitive to controls with congruent head movements.
In human-robot interaction, addressing disparities in action perception is vital for fostering effective collaboration. Our study delves into the integration of explanatory mechanisms during robotic actions, focusing on aligning robot perspectives with the human’s knowledge and beliefs. A comprehensive study involving 143 participants showed that providing explanations significantly enhances transparency compared to scenarios where no explanations are offered. However, intriguingly, lower transparency ratings were observed when these explanations considered participants’ existing knowledge. This observation underscores the nuanced interplay between explanation mechanisms and human perception of transparency in the context of human-robot interaction. These preliminary findings contribute to emphasize the crucial role of explanations in enhancing transparency and highlight the need for further investigation to understand the multifaceted dynamics at play.
We designed a wine recommendation robot and deployed it in a small supermarket. In a study aimed to evaluate our design we found that people with no intent to buy wine were interacting with the robot rather than the intended audience of wine-buying customers. Behavioural data, moreover, suggests a very different evaluation of the robot than the surveys that were completed. We also found that groups were interacting more with the robot than individuals, a finding that has been reported more often in the literature. All of these findings taken together suggest that a novelty effect may have been at play. It also suggests that field studies should take this effect more seriously. The main contribution of our work is in identifying and proposing a set of indicators and thresholds that can be used to identify that a novelty effect is present. We argue that it is important to focus more on measuring attitudes towards robots that may explain behaviour due to novelty effects. Our findings also suggest research should focus more on verifying whether real user needs are met.
Amidst the Covid-19 pandemic, distance learning was employed on an unprecedented level. As the lockdown measures have eased, it has become a parallel option alongside traditional in-person learning. Nevertheless, the utilization of basic videoconferencing tools such as Zoom, Microsoft Teams, and Google Meet comes with a multitude of constraints that extend beyond technological aspects. These limitations are intricately linked with human behavior, psychology, but also pedagogy, drastically changing the interactions that take place during learning. Telepresence robots have been widely used due to their advantages in enhancing a sense of in-person. To investigate the opportunities, the impact, and the risks associated with the usage of telepresence robots in an educational context, we conducted an experiment in a real setting, in the specific use case of a design school and a project-based class. We are interested in the experience of a classroom and the relationships between a distance student, his/her peers, and the professor/instructor. This study employed two types of robots: a Kubi robot (a semi-static tablet-based system) and a Double robot (a mobile telepresence robot). The primary objective was to ascertain the perceptions and experiences of both remote and in-person students during their interaction with these robots. The results of the study demonstrate a marked preference among students for the Double robot over the Kubi, as indicated by their feedback.
This study explores the influence of teaching methods, task complexity, and user characteristics on perceptions of teachable robots. Analysis of responses from 138 participants reveals that both Teaching with Evaluative Feedback and Teaching through Preferences were perceived as equally user-friendly and easier to use compared to the non-interactive condition. Additionally, Teaching with Evaluative Feedback enhanced robot responsiveness, while Teaching with Preferences yielded results similar to the passive Download condition, suggesting that the degree of interactivity and human guidance in the former may not substantially impact user perceptions. Personality traits, particularly extraversion and intellect, shape teaching method preferences. Task complexity influenced the perceived anthropomorphism, control, and responsiveness of the robot. Notably, the classification task led to higher anthropomorphism, control, and responsiveness scores. Our findings emphasise the importance of task design and the need of tailoring teaching methods to the user’s personality to optimise human-robot interactions, particularly in educational contexts. Project website: https://sites.google.com/view/teachable-robots .
Traditional robotic systems require complex implementations that are not always accessible or easy to use for Human-Robot Interaction (HRI) application developers. With the aim of simplifying the implementation of HRI applications, this paper introduces a novel real-time operating system (RTOS) designed for customizable HRI - RoboSync. By creating multi-level abstraction layers, the system enables users to define complex emotional and behavioral models without needing deep technical expertise. The system's modular architecture comprises a behavior modeling layer, a machine learning plugin configuration layer, a sensor checks customization layer, a scheduler that fits the need of HRI, and a communication and synchronization layer. This approach not only promotes ease of use without highly specialized skills but also ensures real-time responsiveness and adaptability. The primary functionality of the RTOS has been implemented for proof of concept and was tested on a CortexM4 microcontroller, demonstrating its potential for a wide range of lightweight simple-to-implement social robotics applications.
Spoken language is the most natural way for a human to communicate with a robot. It may seem intuitive that a robot should communicate with users in their native language. However, it is not clear if a user's perception of a robot is affected by the language of interaction. We investigated this question by conducting a study with twenty-three native Czech participants who were also fluent in English. The participants were tasked with instructing the Pepper robot on where to place objects on a shelf. The robot was controlled remotely using the Wizard-of-Oz technique. We collected data through questionnaires, video recordings, and a post-experiment feedback session. The results of our experiment show that people perceive an English-speaking robot as more intelligent than a Czech-speaking robot (z = 18.00, p-value = 0.02). This finding highlights the influence of language on human-robot interaction. Furthermore, we discuss the feedback obtained from the participants via the post-experiment sessions and its implications for HRI design.
As more and more social robots are being used for collaborative activities with humans, it is crucial to investigate mechanisms to facilitate trust in the human-robot interaction. One such mechanism is humour: it has been shown to increase creativity and productivity in human-human interaction, which has an indirect influence on trust. In this study, we investigate if humour can increase trust in human-robot interaction. We conducted a between-subjects experiment with 40 participants to see if the participants are more likely to accept the robot's suggestion in the Three-card Monte game, as a trust check task. Though we were unable to find a significant effect of humour, we discuss the effect of possible confounding variables, and also report some interesting qualitative observations from our study: for instance, the participants interacted effectively with the robot as a team member, regardless of the humour or no-humour condition.
