Waffler: A Constraint-Directed Approach to Intelligent Agent Design
John Anderson · 1997
andersj @ cs.umanitoba.ca One of the most significant applications of constraint-directed reasoning to the design of intelligent agents is its extremely broad and general mechanisms for knowledge representation. This paper explores the use of constraint-directed reasoning as a foundation for real-time intelligent agents in complex domains, and describes an architecture that employs constraint-directed reasoning to deal with both high and low-level aspects of agent functionality. It represents both a summary of work performed to date, as well as work in progress. 1.