Artificial Intelligence for Social Robots Interacting with Humans in the Real World.

Abstract: Nowadays with the advance of artificial intelligence (AI) and robotics, there is a growing move to use AI-enabled robots for human-robot interaction to do wide range of diverse tasks combining the best abilities of human and robots to achieve the success. The problem is that social interaction is very complex as its dependents not only on physics but also on a multitude of psychological, sociological and other factors. This makes reasoning and learning about it incredibly challenging. Therefore, nowadays most of the HRI researchers are focus on how to design interactions in specific situations, but if we envision service robots that will be able to collaborate efficiently with humans, these robots need to learn how to interact in general with humans in a socially suitable manner. Furthermore, this workshop is related with a trilateral project hold by the co-organizers that is called AI4HRI. We are researchers of the trilateral project Artificial Intelligence for Human-Robot Interaction (AI4HRI) at LAAS, CNRS, France with Aurélie Clodic and Rachid Alami (https://www.laas.fr); HRI Lab., Kyoto University, Japan with Takayuki Kanda (https://www.robot.soc.i.kyoto-u.ac.jp/en/); and IAI, Bremen University, Germany with Michael Beetz (https://ai.uni-bremen.de/). The goal of the AI4HRI project is to combine joint action planning and execution, social interaction learning, and knowledge reasoning and representation into a unified architecture for human-robot interaction. In this way, more intelligent robot's behaviors for human-robot collaboration to accomplish joint goals can be achieved. Objectives: The main objective of the workshop is to bring together a community of researchers interested in making human-robot social interaction more competent with the help of artificial intelligence that in the future can make new collaborations to improve the research in this field by combining their best abilities, as we have in the AI4HRI project. Other objectives of this workshop are: • To exchange the most recent works in this field of several invited speakers, all of them have interesting works combining AI with HRI. • To talk more closely with experienced researchers in this field, by asking questions or exchanging ideas. • To have interesting discussions about our vision of the future of Artificial Intelligence for Social Robots collaborating with humans from different points of view. We envision two panel discussions of 1h or a huge panel discussion of 2h. Where we can discuss our vision of the future for AI for HRI, including multiple interactions between different robots and humans (with distinct roles) and the possibility of both short- and long-term interactions. • To hold a poster session where other researchers can show some interesting works in the field, selected from the submitted papers to the workshop. • To spread the knowledge acquired through this workshop to as many people as possible. If the speakers also agree, it would be good to include videos of all the talks on the web. If the conference allows, it would be good to have a hybrid workshop (face-to-face and online) to allow as many people to participate as possible. • To support gender equality by having a (near) equal number of male and female speakers in this field. We are also trying to invite people from several countries: USA, Europe, GB, and Japan; to have more diversity of cultures. • To deliver our main message to all the targeted audience - that HRI should be combined with AI to obtain more capable, useful, humanlike robot behaviors. With AI, robots can adapt to different situations based on previous experience, learn new interaction behaviors or information from scratch to improve future human-robot collaboration, work together with humans to accomplish common goals, evaluate their behaviors to improve the ways to fulfil these goals, consider the behaviors and interests of their human partners to involve them in their objectives, etc. Topics of interest: Our general topic for the workshop is Artificial Intelligence for Social Robots Interacting with Humans in the Real World, and the workshop sub-topics include, but are not limited to: • Human-robot joint action planning and execution • Knowledge representation and reasoning for HRI • Learning of social interaction in HRI tasks • Robots interacting with multiple humans and multiple robot HRI • Applications of mobile service robots in social environments with human interaction • Evaluation of the quality of human-robot interaction • Human-robot collaboration models to achieve better results more quickly • Design of robots with humanlike behavior and cognition for HRI • Design more predictable or humanlike robot behaviors for HRI • Facilitate mutual recognition of behaviors between robots and humans • Short- and long-term HRI