A metamorphic internal image of autonomous agents: Emergence of behavior

https://doi.org/10.1016/S0921-8890(99)00013-5Get rights and content

Abstract

This study presents an autonomous agent to be able to behave appropriately in a complex environment with unsophisticated sensors. To satisfy this aim, an internal cognitive space named Metamorphic Internal Image Space with learning mechanism is introduced. The Metamorphic Internal Image Space integrates the sensory inputs and the temporal change of the space, so that the agent behaves relying only on the image. The agent is considered to have the following two major characteristics:

  • 1.

    1. It can learn a chain of actions considered as a behavior.

  • 2.

    2. It can move with/without the change of sensory inputs.

Computational experiments show an emergence of behaviors.

References (12)

  • S. Harnad

    The symbol grounding problem

    Physica D

    (1990)
  • P. Maes

    Situated agents can have goals

  • R.A. Brooks

    Intelligence without representation

    Artificial Intelligence

    (1991)
  • L. Chrinsman

    Reinforcement learning with perceptual aliasing: The Perception Distinctions Approach

  • J. Hakura et al.

    Affordance in autonomous robot: grounding mechanism of sensory inputs

  • J.H. Holland et al.

    Induction: Process of Inference, Learning, and Discovery

    (1986)
There are more references available in the full text version of this article.

Cited by (1)

View full text
Copyright © 1999 Published by Elsevier B.V.