The Workshop BAILAR  (Behavior Adaptation and Learning for Assistive Robotics) will be held

in conjunction with the 32nd International Conference on Robots & Human Interactive Communication (ROMAN 2023), 

28- 31 August, 2023, Busan (Korea).

SCIENTIFIC CONTEXT and OBJECTIVES

Mutual affective understanding is integral for achieving the so-called Theory of Mind (ToM) allowing successful, acceptable and intelligent social human-robot interaction (HRI).

Corresponding concepts rely on new paradigms of robotic control systems that consider not only the possibility for a robot to interpret humans’ observable behaviors and internal/emotional states with the aim of anticipating and adapting to their subsequent reactions, but also to facilitate humans in the interpretation and anticipation of the robot’s state, intentions, and future actions through legible behavior designed around the emotional dimension of the communication. It is crucial to endow a robot with the ability to maintain a model of how human partners perceive the world and the robot itself, so as to be able to understand them better and to generate behaviors that are understandable by them. In addition to being able to detect and interpret humans' basic affective responses so as to adapt their behavior accordingly, robots should also be endowed with the ability to show affective and social responses in a legible way for humans to allow acceptability and efficient human-machine communication. This is especially desirable in the field of Socially and Assistive Robotics (SAR), where interaction often takes place with disabled or vulnerable people. Moreover, robots can help decrease the workload and physical contact of healthcare specialists, which would be beneficial to overcome the situations where elderly or vulnerable people are at risk of getting infected. The need to ensure the positive feelings and acceptance of people, while providing them with the necessary assistance, has outlined the utility of intelligent and empathic socially assistive robots. In these contexts, it is of paramount importance to consider the effect of verbal and non-verbal emotional social cues of the robot on the affective state of the user with whom it is interacting. On the one hand, a robot can be employed to provide assistance to individuals with disability (e.g., cognitive or physical  rehabilitation exercises) with the aim of increasing their health and improving their quality of life. On the other hand, its presence and actions could also potentially trigger negative emotions such as stress or discomfort if not tailored to the particular individual’s needs and expectations. This may cause serious difficulties by negatively impacting the users' health and, in turn, achieving a counter-productive result. In this context, robots could use the affect-sensing capability to learn and adapt their behavior to be more comfortable for individuals and to achieve a higher degree of learnability and acceptability through inclusive interaction. 

This edition of BAILAR Workshop aims at putting the mutual understanding of affective/emotional states of robots and humans at the crossroads with learning and user adaptation within HRI. Methodologies and technologies adopted for detecting and adapting to users’ mental states, emotions and dispositions during HRI will be presented and discussed. Experimental protocols and results could also outline possible effects of gender, age, personality and pathology on robot perception from an emotional and affective point of view, as well as ethical considerations regarding learning and using personal data within assistive applications.

Target Audience

This workshop is intended as a forum for a broad audience, composed of roboticists, psychologists, computer scientists, social and ethics scientists, aiming at discussing the role of adaptation, learning, emotional communication, and mutual affective understanding in HRI for assistive real-world applications, and their potential impact on users’ acceptance, especially in terms of usability, efficiency, empathy and their emotional responses. Based on previous editions we can foresee about 80 attendees.

List of Topics

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

• Online adaptive behavior

• Acceptability and personalization in HRI

• User and behavior modeling for adaptation

• Emotion and intention recognition

• Multimodal interfaces for emotion recognition

• Affective and emotional robots

• Empathy in robotics

• Mutual affective communication

• Affective computing in SAR

• Internal state monitoring

• Detection of verbal and non-verbal behavioral cues

• Lifelong (or continual) learning and adaptation methods in HRI

• Short-term personalization in HRI

• Incremental and online learning in HRI

• Robot behavior explainability and legibility

• Theory of Mind in Robotics

• Understandable Communication

• Ethical Aspects of Affective HRI communication and personalization in SAR

• Evaluation metrics for adaptive robot behavior                                                                                                                                                                                                      

NEWS

• The Workshop will be held on the 31st of August, 2023 (Paradise Hotel, Busan, Korea, from 13:10 to 16:00 local time (from 6:10 to 09:00 CEST).