BAILAR 2021

Welcome to the Workshop on Behavior Adaptation, Interaction and Learning for Assistive Robotics - BAILAR!

This workshop is held in conjunction with the IEEE International Conference on Robot and Human Interactive Communication (RO-MAN 2021) conference. Both the workshop and main conference will be held virtually. The Workshop will be held on August 8th (8:00-12:00 MST / 16:00 -20:00 CET).

Statement of Objectives

Endowing robots with learning and behavioral adaptation abilities is a key objective for enabling natural and efficient human-robot interaction, especially in the areas of assistive robotics. As one of the novels of Asimov pointed out in The Complete Robot (1982), a robot that can learn from experience might make mistakes before being able to correctly adapting to the human and the environment. Therefore, it is crucial to design a robot that is efficient, accepted and secure for humans.

Efficiency can be achieved by robots if they are capable to adapt their behaviors on-the-fly according to the interaction with the user. Interactions based on a set of predefined behaviors can cause a decrease of the user engagement over time. In contrast to "one-size-fits-all" approach, robots are requested to offer assistance by tailoring their behaviors to user's preferences, personality and by recalling past experience (so that they do not start each new interaction from scratch). The constant adaptation and personalization require learning a model of human behavior and integrating this model into the decision-making algorithm of the robot. On-the-fly efficient adaptation also requires model-free learning abilities to cope with the high stochastic nature of the environments, and the changing of the human’s attitude, their motivations, and the unpredictability of the interactions.

Acceptability can be achieved by designing robots in accordance with a human-centred approach considering different aspects, such as usefulness, enjoyment, sociability, companionship and perceived behavioral control. Additional aspects, such as ethical considerations and psychological/cognitive aspects, must be considered when robots are interacting with vulnerable populations (older adults, people with cognitive and physical impairment, children with autism, etc).

Safety and security can be achieved by addressing people’s well-being during close interactions with a robot, and their concerns related to cybersecurity attacks, and privacy violations. People are still worried of deploying social robots for assistive purposes even if their potential and usefulness has been largely recognized. Consequently, safety and people’s perception of safety are fundamental requirements when designing robots to be used to interact with people.

These crucial aspects can be enhanced if we consider a natural two-way human-robot communication. They require robots that are able to detect and understand social cues (verbal and non-verbal), human activities, intentions, and internal states, and to show similar behaviors to humans.

Target Audience

This workshop is intended as a forum for a broad audience, which spans from social and assistive robotics, to user-profiling, machine learning and robot behavior control. It will constitute a unique opportunity to gather roboticists and cognitive and computer scientists to present a variety of current approaches aiming at endowing social assistive robots with adaptive and learning capabilities, enhanced cognitive and social abilities, and discuss their potential to meet these criteria. This will permit us to analyze the current state of the field and of its possible real-world applications in assistive robotics.

Topics of interest include, but are not limited to, the following:

• Social and Assistive Robotics

• User Profiling and Personalization in HRI

• On-line Learning and Adaptation

• Activity, Intention and Emotion Recognition

• Cognitive Robotics Architectures in HRI

• Decision Making and Planning

• Multimodal Communication in HRI

• Mutual Affective Understanding in HRI

• Psychophysiological Monitoring and Assessment

• Acceptability, Trustworthiness and Ethics in HRI

• Security and safety in HRI

• Brain-Computer interfaces for HRI

• Theory of Mind in Robotics

• Metrics for HRI Quality

• Explainability and Transparency in HRI

News

• Extended Deadline for Paper Submission: July 10th, 2021

• Professor Yiannis Demiris, Imperial College of London, UK, has been confirmed as an invited speaker.

• Professor Séverin Lemaignan, University of the West of England, UK, has been confirmed as an invited speaker, particularly covering the topic of Socially-driven autonomous Robots.

• Professor Alessandro Di Nuovo, Sheffield Hallam University, UK, has been confirmed as an invited speaker.

• Professor Angelo Cangelosi, University of Manchester, UK, has been confirmed as an invited speaker, particularly covering the topic of Developmental Robotics for Language Learning, Trust and Theory of Mind.

• Selected papers will be invited to submit an extended/revised version on a high impact Journal Special Session.

• The Workshop will be held on August 8th.