What can i do with this?: finding possible interactions between characters and objects

ABSTRACT

Virtual environments are often populated by autonomous synthetic agents capable of acting and interacting with other agents as well as with humans. These virtual worlds also include objects that may have different uses and types of interactions. As such, these agents need to identify possible interactions with the objects in the environment and measure the consequences of these interactions. This is particularly difficult when the agents never interacted with some of the objects beforehand. This paper describes SOTAI - Smart ObjecT-Agent Interaction, a framework that will help agents to identify possible interactions with unknown objects based on their past experiences. In SOTAI, agents can learn world regularities, like object attributes and frequent relations between attributes. They gather qualitative symbolic descriptions from their sensorial data when interacting with objects and perform inductive reasoning to acquire concepts about them. We implemented an initial case study and the results show that our agents are able to acquire valid conceptual knowledge.

REFERENCES

• 1: Abaci, T., Ciger, J., Thalmann, D.: "Action Semantics in Smart Objects" In: Proceedings of the Workshop towards Semantic Virtual Environments, Villars, Switzerland pp.121--126 (2005)
• 2: Abaci, T., Ciger, J., Thalmann, D.: "Planning with Smart Objects" In: WSCG (Short Papers) 2005: 25--28
• 3: Marc Cavazza , Simon Hartley , Jean-Luc Lugrin , Mikael Le Bras, Qualitative physics in virtual environments, Proceedings of the 9th international conference on Intelligent user interfaces, January 13-16, 2004, Funchal, Madeira, Portugal [doi>10.1145/964442.964454]
• 4: Marc Cavazza , Fred Charles , Steven J. Mead, Agents' Interaction in Virtual Storytelling, Proceedings of the Third International Workshop on Intelligent Virtual Agents, p.156-170, September 10-11, 2001
• 5: Cavazza, M., Charles, F., Mead, S. J., "AI-based Animation for Interactive Storytelling", Proceedings of Computer Animation, 2001. IEEE Computer Society Press.
• 6: Paul R. Cohen , Marc S. Atkin , Tim Oates , Carole R. Beal, Neo: learning conceptual knowledge by sensorimotor interaction with an environment, Proceedings of the first international conference on Autonomous agents, p.170-177, February 05-08, 1997, Marina del Rey, California, United States [doi>10.1145/267658.267693]
• 7: Paul R. Cohen , Tim Oates , Carole R. Beal , Niall Adams, Contentful mental states for robot baby, Eighteenth national conference on Artificial intelligence, p.126-131, July 28-August 01, 2002, Edmonton, Alberta, Canada
• 8: Cos-Aguilera, I., Cañamero, L., Hayes, G. M., "Learning Object Functionalities in the Context of Behaviour Selection", in Proceedings of the 4th British Conference Towards Intelligent Autonomous Robots (TIMR'03), August 2003
• 9: Cos-Aguilera, I., Cañamero, L., Hayes, G. M., "Using a SOFM to learn Object Affordances", In Proceedings of the 5th Workshop of Physical Agents (WAF'04), Girona, Catalonia, Spain. March, 2004. DRAFT
• 10: Forbus, K., "Qualitative reasoning". CRC Hand-book of Computer Science and Engineering. 1996, CRC Press.
• 11: Gibson, J. J. (1979). "The Ecological Approach to Visual Perception". Boston: Houghton Mifflin.
• 12: Luiz M. G. Gonçalves , Marcelo Kallmann , Daniel Thalmann, Programming Behaviors with Local Perception and Smart Objects: An Approach to Solve Autonomous Agents Tasks, Proceedings of the 14th Brazilian Symposium on Computer Graphics and Image Processing, p.184-191, October 15-18, 2001
• 13: Gonçalves, L. M., Kallmann, M., and Thalmann, D., "Defining Behaviors for Autonomous Agents based on Local Perception and Smart Objects" In Computer & Graphics, 26(6):887 897, December 2002
• 14: Kallmann, M., Thalmann, D., "Modeling Objects for Interactive Tasks" In: EGCAS'98 - 9th Eurographics Workshop on Animation and Simulation, Lisbon (1998)
• 15: Kallmann, M., "Object Interaction in Real-Time Virtual Environments", DSc Thesis number 2347, Swiss Federal Institute of Technology (EPFL), January 2001.
• 16: Levison L., Geib C., Moore M., "Sodajack: an architecture for agents that search for and manipulate objects" In: Technical Report MS-CIS-94-16, University of Pennsylvania, Dept. of Computer and Information Science, Philadelphia, PA, 1994.
• 17: Levison, L., "Connecting Planning and Acting: Towards an Architecture for Object-Specific Reasoning" In: Dissertation proposal, 1994.
• 18: McGrenere J., Ho, W., "Affordances: Clarifying and evolving a concept". In Proceedings of graphics interface (2000, May) (pp. 179--186), Montreal, Canada, Retrieved September 1, 2003, from http://www.graphicsinterface.org/proceedings/2000/177/
• 19: Viezzer, M. "The Autonomous Acquisition of Affordance Concepts: Exploring a Neighbourhood in Design Space", Technical Report CSRP-03-3, School of Computer Science, University of Birmingham, October 2003.
• 20: Viezzer, M., Nieuwenhuis, C. M. "Learning affordance concepts: some seminal ideas", In IJCAI05 Workshop on Modeling Natural Action Selection.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.6 Learning

General Terms:
Algorithms, Experimentation, Human Factors

Keywords:
believable qualities, human-like, learning agents, lifelike, synthetic agents