| 1. |
- Calvo Barajas, Natalia, 1988-
(författare)
-
Exploring Multidimensional Trust : Shaping Child-Robot Creative Collaborations in Education
- 2023
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- As trust plays a pivotal role in maintaining long-term interactions between children and robots, it is vital to comprehend how children conceptualise trust and the factors influencing their trust in robots. This thesis examines the impact of social robots' behaviours and attributes on children's trust, relationship formation, and task performance in collaborative educational scenarios. A systematic review of child-robot interaction (cHRI) literature identified two primary dimensions of trust: social trust and competency trust. The literature suggests a lack of consensus about how different robot behaviours and attributes affect these two dimensions of trust, as evidence points to different directions. To address these gaps, a collaborative storytelling game was developed to facilitate interactions between children and social robots, aiming to study trust dynamics and enhance learning by fostering children's creativity. The research also examined the impact of robot-related factors, such as behaviour and appearance, on children's interactions with robots. Empirical evidence suggests that while making robots look and behave more like humans is critical for competency trust and task performance, lower human-like attributes are more crucial for developing social trust and relationship formation with robots. Other factors, like time, provide insights into children's trust dynamics. Thus, this thesis explores the role of repeated interactions with artificial agents, indicating that children's competency trust in robots changes over time. This thesis offers significant contributions to the cHRI community. Firstly, it demonstrates that trust is a multidimensional construct that is complex to capture, highlighting the need for reliable, objective measures tailored to the task and intended trust dimension. Secondly, it emphasises the importance of balancing human likeness with social robots when collaborating with children in educational scenarios. Lastly, it proposes that to sustain trustworthy long-term interactions in education; social robots should adapt their behaviour to provide scaffolding, as children will be more inclined to rely on them for learning support as time progresses.
|
|
| 2. |
- Gao, Yuan
(författare)
-
Machine Behavior Development and Analysis using Reinforcement Learning
- 2020
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- We are approaching a future where robots and humans will co-exist and co-adapt. To understand how can a robot co-adapt with humans, we need to understand and develop efficient algorithms suitable for our interactive purposes. Not only it can help us to advance the field of robotics but also it can help us to understand ourselves. A subject Machine Behavior, proposed by Iyad Rahwan in a recent Science article, studies algorithms and the social environments in which algorithms operate. What this paper's view tells us is that, when we would like to study any artificial robot we create, like natural science, a two-step method based on logical positivism should be applied. That is, we need to, on one hand, provide a complicated theory based on logical deduction, and on another hand, empirically setup experiments to conduct.Reinforcement learning (RL) is a computational model that helps us to build a theory to explain the interactive process. Integrated with neural networks and statistics, the current RL is able to obtain a reliable learning representation and adapt over interactive processes. It might be one of the first times that we are able to use a theoretical framework to capture uncertainty and adapt automatically during interactions between humans and robots. Though some limitations are observed in different studies, many positive aspects have also been revealed. Additionally, considering the potentials of these methods people observed from related fields e.g. image recognition, physical human-robot interaction and manipulation, we hope this framework will bring more insights to the field of robotics. The main challenge in applying Deep RL to the field of social robotics is the volume of data. In traditional robotics problems such as body control, simultaneous localization and mapping and grasping, deep reinforcement learning often takes place only in a non-human environment. In such an environment, the robot can learn infinitely in the environment to optimize its strategies. However, applications in social robotics tend to be in a complex environment of human-robot interaction. Social robots require human involvement every time they learn in such an environment, which leads to very expensive data collection. In this thesis, we will discuss several ways to deal with this challenge, mainly in terms of two aspects, namely, evaluation of learning algorithms and the development of learning methods for human-robot co-adaptation.
|
|
| 3. |
- Paetzel-Prüsmann, Maike, 1991-
(författare)
-
The Novelty in the Uncanny : Designing Interactions to Change First Impressions
- 2020
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- In 1970, Japanese researcher Masahiro Mori published a seminal paper where he hypothesized that robots that appear human-like but are still distinguishable from being human would not attract people towards them, but instead cause an uncanny sensation. This phenomenon, known as the uncanny valley effect, has been widely studied within the social robotics community, and a multitude of experiments have since been conducted supporting Mori's hypothesis. The specifics of a robot's appearance and behavior leading to such an uncanny sensation, however, remain an open research question and require further study. These gaps in the causal relationship between uncanny feelings and a robot's design have lead uncanniness being increasingly used to explain any lack of enthusiasm towards robots, both in the scientific community and the general public. It is then often implicitly assumed that uncanny feelings towards a robot have damaging consequences for long-term human-robot interaction. Most empirical studies on the subject, however, focus on still images or short video clips of robots and participants are only exposed to these stimuli for small frames of time. The current literature on the uncanny valley does not thus allow a conclusion to be drawn about the persistence of uncanny feelings. This thesis addresses this gap in the body of knowledge by implementing interactive scenarios and performing a series of empirical investigations to study the development of people's uncanny feelings towards social robots over the course of one or several such interactive encounters with them. The findings suggest that novelty plays an important role in the feeling of uncanniness: Merely interacting with a robot for a brief period and thus giving human observers access to the robot's full behavioral stream lowers their rating of uncanny feelings towards the robot as compared to how they perceive it at first sight. Furthermore, repeated interactions with a robot can further lower uncanny impressions. These results contribute to the field of human-robot interaction, as they posit that increased exposure may result in limited feelings of uncanniness. This, in turn, potentially reduces the impact of uncanny feelings on long-term interactive encounters with robots. Instead of focusing on reducing the elicitation of uncanny first impressions, it may thus be more sustainable to further study how interactions can help people efficiently get to know a robot and overcome their initial reluctance towards it.
|
|
| 4. |
- Wallkötter, Sebastian, 1993-
(författare)
-
Transparency Mechanisms in HRI : Improving an observer’s understanding of social robots
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- During an interaction between a robot and its user, a robot may sometimes do things that the user finds unintuitive. This often happens because the user does not understand the robot’s intent, state, or policy well enough. In such cases, users would benefit if the robot had the ability to reveal this hidden information; a property which is called transparency. Transparency is also desirable because it helps robots comply with ethical guidelines, makes interactions more robust, and increases users' trust in the robot. Here, we investigate how robots can be made transparent, and our first step towards this is a literature review of the area. After completing the review, we suggest using available robot modalities and information content as features to find suitable technical approaches (frameworks) for transparency which we identify and categorize. In addition, we use these features to break transparency down into more manageable pieces, which we call types of transparency, and also find that situatedness, i.e., if the interaction takes place in a physical or virtual space, changes the effect of the robot’s communications. We then narrow our attention and focus on legibility, a type of transparency that uses movement to communicate the robot’s intent. Here, we investigate when to use which legibility framework, and propose a novel approach to benchmark them. Leveraging these findings, we then propose our own machine-learning-based legibility framework, which is general enough to be able to imitate several existing legibility frameworks and which can learn a user’s expectations from data.
|
|
