Tutorials and Special Sessions

Accepted Tutorials for RO-MAN 06

1. Socio-emotional Interaction with Virtual Humans

Abstract:
As real-time animated characters become almost commonplace, we begin to face the next challenges of making those characters inhabit social contexts, communicate with people, and exhibit behaviors which convey non-verbal information. Recent research has developed cognitive, psychosocial, psycholinguist and affective models for human-like agents. An agent’s “mind” may now be connected to its body (enabling more human-like behaviors by animating expressive and coordinated movements in language, gesture, expression and gaze) and to the mind of social partners (through speech recognition, prosodic analysis, gesture and gaze detection, expression recognition and physiological sensors). This tutorial will present an overview of the psychosocial findings that underlie and motivate the design of such systems, and the techniques and methods necessary for their construction. Applications to interactive training, smarter games, and group simulations will be addressed.

» Detailed Information

2. Coding Video Data in Human-Robot Social Interaction: Behavior and Reasoning

Abstract:
As the field of human-robot interaction matures, researchers and designers are recognizing the need for systematic, comprehensive, and theoretically-grounded methodologies for investigating people’s social interactions with robots. One attractive approach entails the collection of video data in naturalistic or experimental settings. Depending on the research paradigm, subjects may interact socially with a robot around a specified task, in response to specified questions, or in unstructured interactions. With video data in hand, the question then emerges: How does one code and analyze such data? Key challenges entail identifying behavioral segments, identifying relationships among behaviors, accounting for “spontaneous” robot-initiated behaviors, accounting for dialog as part of the interactions, and establishing reliability of the coded system.
This tutorial introduces participants to one established method developed by our laboratory for coding and analyzing video data of people’s social interactions with robots. The coding system was initially developed based on preschool children’s interactions with Sony’s robotic dog AIBO and a stuffed dog (Kahn, Friedman, Perez-Granados, & Freier, 2004) and subsequently applied to older children’s and adolescents’ interactions with AIBO and a live dog (Melson, Kahn, Beck, Friedman, Roberts, & Garrett, 2005). It is currently being used to formulate studies of social interaction with humanoid robots.
An innovative aspect of this tutorial entails – in an interactive workshop format – applying principles from the existing coding system to new video data supplied by one or more of the tutorial participants.

» Detailed Information

3. Cognitive Robotics

Abstract: Cognitive robotics seeks to develop intelligent robust machines, whilst also exploring the nature of cognition and control. Domains of operation are often characterized by dynamic environments that were previously unexplored. Tasks include human assistance, domestic chores, space exploration and mirroring human cognition to give insight into our own mind.

This tutorial will consider advances in sensorimotor coordination from adaptive control to cognitive control. It will also focus on advances on how the sensed environment may be represented, stored and retrieved. Advances in cognitive architectures that give insight into the complete cognitive process will be explored. A wide range of topics from the Sensory Ego Sphere to cognitive control of robots will be introduced as a case study. Cognitive neuroscience models of important aspects of cognition and learning, such as the Working Memory, will be used to contrast various opinions in the field.

An attendee will gain an understanding of the important advances in the field of cognitive robotics including highlights of the latest developments in computational neuroscience and a guide to the fertile areas for future research.

4. Social Interactions with Robot Companions

Abstract: This tutorial is concerned with research issues that are important for the development and study of cognitive robot companions, part of the European project Cogniron (www.cogniron.org). In the context of the Cogniron project a Robot Companion must fulfil two important functions: a) making itself 'useful', i.e. being able to carry out a variety of tasks in order to assist humans in a domestic home environment, and b) behaving socially, i.e. possessing social skills in order to be able to interact with people in a socially acceptable manner. While this very interdisciplinary research into robot companions involves various hard research problems related e.g. to robotics, computer science, psychology and HCI, the tutorial will specifically focus on aspects of interactivity that are crucial for robot companions who are required to operate in human-inhabited environments and interact with people in an acceptable and believable manner. The tutorial will consist of three parts:

• Human-Robot Interaction Research for Robot Companions (presented by Kerstin Dautenhahn, University of Hertfordshire, UK)
• Issues of designing and implementing a control architecture for an interactive robot (presented by Rachid Alami, LAAS, France)
• Integrating dialog in human-robot interactions (presented by Britta Wrede, University of Bielefeld, Germany)

Submissions of Tutorial and Special Session Proposals

Tutorials:

• Title of tutorial
• Tutorial speaker(s), including short CVs
• Motivation/Background (half a page)
• Structure/Overview of tutorial including topics covered
• Links and references relevant to the tutorial
• Any other relevant material

Special Sessions:

• Organizer(s) of special session
• Proposed theme of special session
• One paragraph motivating the relevance of theme to IEEE RO-MAN 06
• List of 4-5 proposed contributions, including planned titles of papers, names and affiliations of authors, and short abstracts of papers