Human-Robot Interaction for Wellbeing

Socially Adaptive Robots

Full-day Hybrid workshop on September 2nd, 2022 as part of the IEEE International Conference on Robot & Human Interactive Communication (RO-MAN 2022)


Robots have found their way into society only for task-oriented goals, such as cleaning or cooking at home, entertaining at malls, and supporting workers in a job environment. Those robots can perform very restrictive and repetitive tasks without involving any human interaction. In recent years, there has been an increasing interest in companion robots that could support people for their entertainment or wellbeing, such as Amazon Astro (i.e., Alexa embedded into a wheel-robot) or the personal robot Jibo (i.e., developed by the MIT Media Lab to connect with people and being a robotic friend). The need to introduce companion robots in clinical (e.g., therapeutic centers, hospitals) and non-clinical (e.g., homes, work environments, malls) contexts have been boosted by the COVID-19 pandemic, where many people were forced to stay at home alone or with their relatives. This has caused many wellbeing issues, such as depression and anxiety, to mention only a few examples, and has resulted in a higher interest in assistive technology to alleviate the everyday burden. Socially Assistive Robots (SARs) are a promising venue to support people in their lives and help improve their wellbeing. However, due to the lack of large scale datasets obtained over longer periods of time SARs are very limited in their capabilities to continuously adapt to each user’s needs and socio-emotionally connect with them. This often increases the risk of perceiving the robots as asocial leading to user disappointment and dissatisfaction, which are crucial in a domain that requires adaptation and socio-emotional behavior to increase robots’ social capabilities to promote wellbeing.

For example, consider a socially assistive robot placed in the university students’ residential rooms. It can assist university students in their everyday life tasks and monitor their wellbeing during stressful periods of their uni life (e.g., mid-terms, finals). If the student is too stressed out or is close to burnout, the robot could perceive the emotional and mental state of the student - detecting the emotion, stress, and depression levels of the student - and adapt to the student’s needs, specifically it could respond empathically trying to support the student conversationally, and promoting mindfulness sessions, or suggesting external support (e.g., “I know that exams are really stressful, but you need to take some time for yourself, would you like to do a meditation session?”). In this situation, if a robot is not able to comfort the student properly (e.g., if the student is stressed out for university exams, the robot keeps reminding him/her about the deadlines), this leads to student’s frustration fueling the risk of perceiving the robot as asocial and reducing the student’s willingness to interact with it.

List of topics

Topics included in the workshop, but not limited to, are the following: