A context-based architecture for general problem solving

  • Authors:
  • R. C. Peterson

  • Affiliations:
  • School of Computing, University of Leeds, Leeds, LS2 9JT, UK.

  • Venue:
  • ICSAB Proceedings of the seventh international conference on simulation of adaptive behavior on From animals to animats
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

Perhaps the most striking feature of the human problem solving ability is its apparent generality. Many animals have well developed path finding and navigational abilities and this suggests a route by which general purpose problem solving skills might have arisen in primates. The idea explored in this paper is that by generalising the notion of 'place' to include contexts defined by factors other than just spatial location a powerful general purpose representational and planning mechanism can be built on the foundations provided by existing models of animal navigation. In the model presented the animat is assumed to have available to it motor primitives that not only allow it to move within and modify its environment but also allow it to change the way in which its sensory system operates, e.g. change its field of view. By selectively directing its attention in this way the animat is able to build up a dynamic multi-scale representation of its environment.