SC17: Affective Computing

Speaker:

Christian Becker-Asano

Disciplines/fields:

Artificial intelligence, virtual reality, robotics, cognitive science, emotion psychology

Duration:

4 sessions

Course content:

1. Affective Computing: What affect is about and why to compute it
2. Virtual humans: How to create believable characters for virtual worlds
3. Social robots: Embodied interactivity for the real world
4. Android robots: Geminoids and beyond


Objectives:

The primary goal of this special course is to make aware the complexity of modelling affect from an artificial intelligence (AI) point of view. We will start with an overview of what the term "affect" subsumes and how such concepts as emotion, mood, and personality have been conceptualized in the fields of cognitive science, psychology, and neuro-biology. Arguments will be given, why such vague concepts are not only of interest to writers of science fiction stories, but also to AI researchers, who aim to create computational systems that interact naturally with humans.

This will naturally lead us to the interesting and challenging question of how to create virtual humans that inhabitate virtual worlds such as those presented in modern computer games. In contrast to other courses, however, we will focus on the effects of emotions in playful interaction--both on the virtual as well as real human's side. We will learn about the problems that arise when trying to acquire data about a human player's emotional experience during gameplay.
Taking the step from the virtual to the real world a number of (humanoid) robots will be introduced, which are designed to interact socially. The problematic assumption that a "humanoid" form necessarily supports "social" interactivity will be questioned and finally we might be convinced that such non-humanoid robots as "Roomba" can be labeled "social" just as well. We will then learn how humanoid robots such as Robovie can affect people only using "laughter" as a para-verbal signal.

Finally, Geminoid HI-1 will be introduced and how it is used to conduct basic research. We will see how such an android robot is perceived and treated by naive people, who don't expect such a robot's presence in a public cafe. This data will then be complemented with a discussion of a series of laboratory studies in relation to the research paradigm of Affective Computing.

Literature:

Session 1:

  • Picard, R. Affective computing: challenges, in: Int. J. Human-Computer Studies, 2003, 59, 55-64
  • Bartneck, C., Integrating the OCC Model of Emotions in Embodied Characters, Workshop on Virtual Conversational Characters: Applications, Methods, and Research Challenges, 2002
  • Becker-Asano, C., WASABI: Affect Simulation for Agents with Believable Interactivity, chapter 2, Faculty of Technology, University of Bielefeld, 2008-> http://www.becker-asano.de/Becker-Asano_WASABI_Thesis.pdf

Session 2:

Session 3: 

  • Dautenhahn, K.; Nourbakhsh, I. & Fong, T., A Survey of Socially Interactive Robots, Robotics and Autonomous Systems, 2003, 42, 143-166
  • Hudlicka, E.; Payr, S.; Ventura, R.; Becker-Asano, C.; Fischer, K.; Leite, I.; Paiva, A. & von Scheve, C., Social interaction with robots and agents: Where do we stand, Where do we go?, Affective Compting and Intelligent Interaction, 2009, 698-703-> http://www.becker-asano.de/ACII2009PanelContrib.pdf
  • Becker-Asano, C.; Kanda, T.; Ishi, C. & Ishiguro, H., How about laughter? Perceived naturalness of two laughing humanoid robots, Affective Computing and Intelligent Interaction, 2009, 49-54-> http://www.becker-asano.de/HowAboutLaughter_ACII2009.pdf

Session 4:

  • Nishio, S.; Ishiguro, H. & Hagita, N., de Pina Filho, A. C. (ed.), Geminoid: Teleoperated Android of an Existing Person, Humanoid Robots, New Developments, I-Tech, 2007, 343-352
  • Becker-Asano, C. & Ishiguro, H. Laughter in Social Robotics - no laughing matter Intl. Workshop on Social Intelligence Design (SID2009), 2009-> www.becker-asano.de/SID09_LaughterInSocialRoboticsCameraReady.pdf

Vita:

Christian Becker-Asano received his doctor's degree (Dr. rer. nat.) from the University of Bielefeld in 2008, for his work on affect simulation for agents with believable interactivity, which he applied to the virtual human Max. He was a Ph.D. student and research assistant in the Artificial Intelligence Group at the University of Bielefeld, Faculty of Technology and a Pre-Doctoral Fellow of the Japan Society for the Promotion of Science (JSPS) during a three month visit in 2005 at the National Institute of Informatics, Tokyo.In his current research, he applies his experience in Affective Computing to the humanoid robots and androids of the Intelligent Robotics and Communication Labs at ATR in Kyoto, Japan, under the guidance of Prof. Hiroshi Ishiguro as a post-doctoral fellow of JSPS and Alexander von Humboldt foundation.

 

 

 

Last update: 21.01.2010, Webadmin