Search

We report on the exploratory design and study of a robot math tutor that can provide feedback on specific errors made by children solving basic addition and subtraction problems up to 100. We discuss two interaction design patterns, one for speech recognition of answers when children think aloud, and one for providing error-specific feedback. We evaluate our design patterns and whether our feedback mechanism motivates children and improves their performance at primary schools with children (N=41) aged 7–9. We did not find any motivational or learning effects of our feedback mechanism but lessons learnt include that the robot can execute our interaction design patterns autonomously, and advanced algorithms for error classification and adaptation to children’s performance levels in our feedback mechanism are needed.

We are developing a social robot that should autonomously interact long-term with pediatric oncology patients. The child and the robot need to get acquainted with one another before a long-term interaction can take place. We designed five interaction design patterns and two sets of robot behaviors to structure a getting acquainted interaction. We discuss the results of a user study (N = 75, 8–11 y.o.) evaluating these patterns and robot behaviors. Specifically, we are exploring whether the children successfully got acquainted with the robot and to what extent the children bonded with the robot. Results show that children effectively picked up how to talk to the robot. This is important, because the better the performance the more comfortable the children are, the more socially attractive the robot is, and the more intimate the conversation gets. The evaluation furthermore revealed that it is important for children, in order to get familiar with the robot, to have shared interests with the robot. Finally, most children did initiate a bond with the robot.

When a robot approaches a person, the chosen trajectory ideally informs the person not only about the robot’s intended target location, but also its intended orientation. However, planning a straight line to the goal location does not guarantee a correct final orientation, potentially causing confusion as the robot eventually rotates towards its unsuspecting target. One method that could remedy this problem is minimum jerk path planning, which results in the smoothest possible path that ends in the pre-specified final orientation. The technique is already widely used in robotic arm motion planning, but existing work is lacking for regular path planning. The aim of the current study is to implement minimum jerk path planning for the Nao robot and to evaluate the potential benefit for human observers to infer the intended target of the robot. Results show that minimum jerk path planning significantly improves people’s recognition of the robot’s destination compared to straight line path planning. Meanwhile, the perceived likeability and human likeness of the robot remain the same, suggesting that implementing smooth robot path planning that includes the final orientation leads to more predictable robot approaching behaviour.KeywordsHuman-aware navigation Path planning Robot intention Human-robot interaction

A quarantine is an effective measure in order to contain a disease and it is needed to be used more often in current times. Quarantine forces people to have minimal to no social contact with other humans for a certain period of time. Past work says this isolation can have a serious psychological impact on people’s lives, which can have dramatic consequences. Research can help find the positive and negative experiences of people in quarantine, in order to determine their needs. But how do people respond to quarantine according to their own self accounts? We look to a video platform as a unique opportunity to explore this question. Robots can be used in times of quarantine so isolation can be maintained. However, these robots should be matching the actual needs of the people in quarantine in order to have an effect. This research will use a content analysis of first-hand accounts of people in quarantine in order to find their experiences and needs. After that, there will be an analysis of robots that are used in times of quarantine. Lastly, these two analyses will be used to find out if the robots match the needs of the people in quarantine. We report on two major components to first-hand social media quarantine accounts: emotional response and procedural explanations provided by detainees, and explore potential reasons for them choosing to share these types of content. On top of that, we report on robots that are mentioned by social media, the tasks that they do, and the needs they fulfill. This research will expand on the current knowledge domain of needs in quarantine and will also add to the knowledge domain of the effectiveness of robots in quarantine.

Improving the quality of life of people with dementia in long-term care facilities is very important and can be achieved by designing engaging activities for the residents. The introduction of social robots for people with dementia has already proven its benefits, and we expected that adding contextual cues to this interaction would enhance the positive engagement of these individuals. A total of five participants took part in a comparison study in which they engaged in a free-play session with the robot PLEO and in a free-play session with the robot PLEO within a jungle-themed context. The gaze and arm/hand behaviors of the participants were measured and were used to determine the level of their engagement. Contrary to our expectations, we found a significant decrease in engagement for the sessions where the context was added to the interaction. Our explanation of this result is that the added cues increased the threshold to interact with the robot, but the elderly were still engaged as spectators.

In many scenarios where robots could assist humans, handover situations are essential. But they are still challenging for robots, especially if these are initiated by the human interaction partner. Human-human handover studies report average reaction times of 0.4 s, which is only achievable for robots, if they are able to predict the object transfer point (OTP) sufficiently early and then adapt to the human movement. In this paper, we propose a hand tracking system that can be used in the context of human initiated handover as a basis for human reaching motion prediction. The OTP prediction implemented is based on the minimum jerk model and combines a static estimation utilizing the human’s initial pose and a dynamic estimation from the current hand trajectory. Results are generated and analyzed for a broad spectrum of human initiated scenarios. For these cases we examine the dynamics of different variants of the proposed prediction algorithm, i.e., how early is a robot’s prediction of the OTP within a certain error range? The tracking delivers results with an average delay, after the initialization, of 0.07 s. We show that the OTP prediction delivers results after 75 % of the movement within a 10 cm precision box.

In this study, we argue that the extent to which a robot is persuasive depends on the way a persuasive message is embedded in the robot’s other behaviors. In the current study, in which a robot serves water at a large public event, we find that the same robot utterance, namely skål (cheers) when serving water, is received very differently depending on whether the robot is oriented to the user or not. In particular, if the robot gazes at the user while saying skål, almost all users drink immediately, whereas only 44.5% of the people drink if the robot is looking elsewhere. Similarly, the effectiveness of a water-related joke as a persuasive means depends on previously established mutual gaze. Thus, gaze and speech behavior have to be coordinated to improve the robot’s persuasiveness in the wild.

The introduction of service robots to our daily life requires adaptation of the current navigation strategies. In the presence of humans, robots must be designed to ensure their safety and comfort. This paper proposes a layered costmap architecture that incorporates social norms to generate trajectories compatible with human preferences. The implemented framework creates a social abstraction of the environment – in the form of an occupancy grid – to plan human-friendly paths. It employs information about individuals in the scene to model their personal spaces. In addition, it uses predicted human trajectories to improve the efficiency and legibility of the robot trajectory. Different simulation scenarios resembling everyday situations have been used to evaluate the proposed framework. The results of the experiments have demonstrated its ability to behave according to social conventions. Furthermore, the navigation system was assessed in real life experiments where it was proved capable of following similar paths to those performed by humans.

This paper is about speaker verification and horizontal localisation in the presence of conspicuous noise. Specifically, we are interested in enabling a mobile robot to robustly and accurately spot the presence of a target speaker and estimate his/her position in challenging acoustic scenarios. While several solutions to both tasks have been proposed in the literature, little attention has been devoted to the development of systems able to function in harsh noisy conditions. To address these shortcomings, in this work we follow a purely data-driven approach based on deep learning architectures which, by not requiring any knowledge either on the nature of the masking noise or on the structure and acoustics of the operation environment, it is able to reliably act in previously unexplored acoustic scenes. Our experimental evaluation, relying on data collected in real environments with a robotic platform, demonstrates that our framework is able to achieve high performance both in the verification and localisation tasks, despite the presence of copious noise.

Care robots are often seen to introduce a risk to human, touch based care. In this study, we analyze care workers’ opinions on robot assistance in elderly services and reflect them to the idea of embodied relationship between a caregiver, care receiver and technology. Our empirical data consists of a survey for professional care workers (n = 3800), including registered and practical nurses working in elderly care. The questionnaire consisted scenarios of robot assistance in care work and in elderly services and the respondents were asked to evaluate whether they see them as desirable. The care workers were significantly more approving of robot assistance in lifting heavy materials compared to moving patients. Generally, the care workers were reserved towards the idea of utilizing autonomous robots in tasks that typically involve human touch, such as assisting the elderly in the bathroom. Stressing the importance of presence and touch in human care, we apply the ideas of phenomenology of the body to understand the envisioned robot-human constellations in care work.

Researching human-robot interaction “in-the-wild” cansometimes require insight from different fields. Experiments that involvecollaborative tasks are valuable opportunities for studying HRI anddeveloping new tools. The following describes a framework for an “in thewild” experiment situated in a public museum that involved a Wizardof OZ (WOZ) controlled robot. The UR10 is a non-humanoid collabora- AQ1tive robot arm and was programmed to engage in a collaborative drawingtask. The purpose of this study was to evaluate how movement by a nonhumanoidrobot could affect participant experience. While the currentframework is designed for this particular task, the control architecturecould be built upon to provide a base for various collaborative studies.