Embodiment without a Physical Body

Source: CiteSeer


This paper describes an experimental platform to model embodied cognitive development in a virtual environment. We have started a research project that aims to realize high-level cognitive functions modeled after the development of infants. We began by designing an experimental platform both for collecting interaction data among a developing system (or an infant) , its environment, and its caregiver, and for performing computer simulations of cognitive development. Rather than a physical robot in the real world, we decided to use a developing system in a simple virtual environment. The developing system thus has no physical body; however, it interacts with the environment and the caregiver, and will learn how to act in the environment and communicate with the caregiver. The basic framework of the development system that we plan to build is also described.

Download full-text


Available from: Kazuo Hiraki, Jan 26, 2013
  • Source
    • "The grounding of concepts on the sensorimotor interaction with the environment eliminates the problems of classical AI (lack of robustness; the lack of access to the semantic content of designer-provided symbols or categories; the confusion between the agent's perspective and the observer's perspective). While embodiment generally implies a real physical body, like those of animals and robots, several studies (Quick, Dautenhahn, Nehaniv, & Roberts, 1999; Riegler, 2002; Oka et al., 2001) have argued that the importance of embodiment is not necessarily given by materiality, but by its special dynamic relation with the environment. This relation can also emerge in environments other than the material world, such as computational ones. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper describes a new simulator, Thyrix, designed for em-bodied cognitive science research. The simulator supports 2D agents having articulated arms, capable of manipulating objects from the en-vironment and also of moving in the environment. The agents have sensors for vision, proprioception, and also tactile sensors distributed on the surface of their body. The simulator supports contact detection and resolution, and friction. It uses a simplified, Aristotelian mechan-ics, but can be easily extended to classical, Newtonian mechanics (and also to 3D). We present reasons for using such a simulator and describe design and implementation characteristics.
    Preview · Article · Sep 2003
  • [Show abstract] [Hide abstract]
    ABSTRACT: The concept "triface", which can provide a unified view of the interactions between human, robot and environment, is proposed and discussed. Traditional concept "interface" mainly focuses on the relationship between two elements, e.g. human and robot, human and computer, human and machine. The triface includes the interactions with the third element: environment, which does not belong to human or robot. The environment can play an important role to the robot-human interactions, especially in the case when a robot has to learn both a human and an environmental model autonomously. The variations of the interactions in the triface are discussed, and we describe the concept which will help the robot to learn how to communicate to the human and how to operate the environment. We also introduce ongoing projects, which are done on the experimental platform developed for the triface researches, and explain the similarity of the framework to the infant developmental processes.
    No preview · Conference Paper · Jan 2003
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Abstract This paper reviews the current state of the art in the research concerning the development of autonomous artificial intelligent agents. First, the meaning of specific terms, like agency, automaticity, autonomy, embodiment, situatedness, and intelligence, are discussed in the context of this domain. The motivations for conducting research in this area are then exposed. We focus, in particular, on the importance of autonomous embodied agents as support for genuine artificial intelligence. Several principles that should guide autonomous agent research are reviewed. Of particular importance are the embodiment and situatedness of the agent, the principle of sensorimotor coordination, and the need for epigenetic development and learning capabilities. They ensure the adaptability, flexibility and robustness of the agent. Several design and evaluation considerations are then discussed. Four approaches to the design of autonomous agents—the subsumption architecture, evolutionary methods, biologically-inspired methods and collective approaches—are presented and illustrated with examples. Finally, a brief discussion mentions the possible role of autonomous,agents as a framework for the study of computational applications of the far-from-equilibrium systems theory. Contents
    Preview · Article · Mar 2003