196,078 research outputs found
A user-centred framework for explainable artificial intelligence in human-robot interaction
No full textIl patrimonio geologico del Parco Naturale Regionale dell'Aveto (Liguria Orientale): interventi di valorizzazione e gestione.
No full textToward Robots' Behavioral Transparency of Temporal Difference Reinforcement Learning with a Human Teacher
No full textThe high request for autonomous human-robot interaction (HRI), combined with the potential of machine learning (ML) techniques, allow us to deploy ML mechanisms in robot control. However, the use of ML can make robots' behavior unclear to the observer during the learning phase. Recently, transparency in HRI has been investigated to make such interactions more comprehensible. In this work, we propose a model to improve the transparency during reinforcement learning (RL) tasks for HRI scenarios: the model supports transparency by having the robot show nonverbal emotional-behavioral cues. Our model considered human feedback as the reward of the RL algorithm and it presents emotional-behavioral responses based on the progress of the robot learning. The model is managed only by the temporal-difference error. We tested the architecture in a teaching scenario with the iCub humanoid robot. The results highlight that when the robot expresses its emotional-behavioral response, the human teacher is able to understand its learning process better. Furthermore, people prefer to interact with an expressive robot as compared to a mechanical one. Movement-based signals proved to be more effective in revealing the internal state of the robot than facial expressions. In particular, gaze movements were effective in showing the robot's next intentions. In contrast, communicating uncertainty through robot movements sometimes led to action misinterpretation, highlighting the importance of balancing transparency and the legibility of the robot goal. We also found a reliable temporal window in which to register teachers' feedback that can be used by the robot as a reward
Humanizing Human-Robot Interaction: On the Importance of Mutual Understanding
No full textIn conjunction with what is often called the industry 4.0, the new machine age, or the rise of the robots, the authors of this paper have each experienced the following phenomenon. At public events and roundtable discussions, among our circles of friends, or during interviews with the media, we are asked on a surprisingly regular basis: "How must humankind adapt to the imminent process of technological change? What do we have to learn in order to keep pace with the smart new machines? What new skills do we need to understand the robots?".Paper published in: IEEE Technology and Society Magazine ( Volume: 37, Issue: 1, March 2018). Published version: https://ieeexplore.ieee.org/abstract/document/830714
Trust and Social Engineering in Human Robot Interaction: Will a Robot Make You Disclose Sensitive Information, Conform to Its Recommendations or Gamble?
No full textRobots such as information security and overtrust in them are gaining increasing relevance. This research aims at giving an insight into how trust toward robots could be exploited for the purpose of social engineering. Drawing on Mitnick's model, a well-known social engineering framework, an interactive scenario with the humanoid robot iCub was designed to emulate a social engineering attack. At first, iCub attempted to collect the kind of personal information usually gathered by social engineers by asking a series of private questions. Then, the robot tried to develop trust and rapport with participants by offering reliable clues during a treasure hunt game. At the end of the treasure hunt, the robot tried to exploit the gained trust in order to make participants gamble the money they won. The results show that people tend to build rapport with and trust toward the robot, resulting in the disclosure of sensitive information, conformation to its suggestions and gambling
Visuo-haptic exploration for multimodal memory
No full textWhen faced with a novel object, we explore it to understand its shape. This way we combine information coming from different senses, as touch, proprioception and vision, together with the motor information embedded in our motor execution plan. The exploration process provides a structure and constrains this rich flow of inputs, supporting the formation of a unified percept and the memorization of the object features. However, how the exploration strategies are planned is still an open question. In particular, is the exploration strategy used to memorize an object different from the exploration strategy adopted in a recall task? To address this question we used iCube, a sensorized cube which measures its orientation in space and the location of the contacts on its faces. Participants were required to explore the cube faces where little pins were positioned in varying number. Participants had to explore the cube twice and individuate potential differences between the two presentations, which could be performed either haptically alone, or with also vision available. The haptic and visuo-haptic (VH) exploratory strategies changed significantly when finalized to memorize the structure of the object with respect to when the same object was explored to recall and compare it with its memorized instance. These findings indicate that exploratory strategies are adapted not only to the property of the object to be analyzed but also to the prospective use of the resulting representation, be it memorization or recall. The results are discussed in light of the possibility of a systematic modeling of natural VH exploration strategies
Predicted sensory feedback derived from motor commands does not improve haptic sensitivity
No full text
- …
