We present a new architecture for robot control rooted in notions from Brooks’ subsumption architecture and extended to include an internal representation which matures as it experiences the world. Our architecture is based on the Copycat program of Mitchell and Hofstadter, a model of fluid representation whose details we discuss. We show how our architecture develops a representation of its environment through a continuing interaction with it. The architecture is founded on a dynamical systems interpretation of representation and demonstrates the importance of the use of “embodiment”. It reflects a constructivist epistemology, with the robot designed to utilize its environment in its exploration.

EX: This chapter proposed a constructivist rapprochement to address the shortcomings found in the rationalist, empiricist, and pragmatist traditions. It was argued that a survival-based tension exists between the expectations of the perceiving agent and perceived information. The agent’s expectations can be characterized by Kant’s, Bartlett’s, or Piaget’s schemas that are either reinforced or recalibrated as new information is perceived. Friston (2009) refers to this phenomenon as free energy minimization; Piaget (1970) describes it as continuing to move towards a state of equilibration. A set of five assumptions and 8 follow-on conjectures were proposed to capture this active subject perception dialectic. The set of conjectures included characterizing the meta-concepts of knowledge, meaning, and truth.

In recent years the Artificial Intelligence research group at the University of New Mexico have considered several areas of problem solving in interesting and complex domains. These areas have ranged from the low level explorations of a robot tasked to explore, map, and use a new environment to the development of very sophisticated control algorithms for the optimal use of particle beam accelerators. Although the results of our research have been reflected in computer-based problem solvers, such as the robot discovering and mapping out its world, these computational tasks are in many ways similar to expert human performance in similar tasks. In fact, in many important ways our computer-based approach mimics human expert performance in such domains. This paper describes three of these task domains as well as the software algorithms that have achieved competent performances therein. We conclude this paper with some comments on how software models of a domain can elucidate aspects of intellectual performance within that context. Furthermore, we demonstrate how exploratory problem solving along with model refinement algorithms can support a constructivist epistemology.

Key words: artificial intelligence, constructivist epistemology, human complex problem solving, model refinement, diagnostic reasoning, intelligent control.