Authors: Oksana Arnold and Klaus P. Jantke.
Email: jantke@dfki.de
Source: New Generation Computing Vol. 15, No. 1, 1997, 27 - 58.
Abstract. This is our most comprehensive paper describing the inductive plan generation approach within the overall concept of knowledge-based process supervision and control. Therapy plan generation for complex dynamic processes is essentially inductive. As a consequence, inductive learning can draw benefit from therapy plan generation and vice versa. The paper focusses on this mutual stimulation. Consistency of plan generation as well as versions of incrementally synthesizing plans are discussed in some more detail. The paper contains the theoretical foundations in logic and graph theory as well as their applications. There is presented a couple of planning algorithms.
©Copyright 1998 Ohmsha,Ltd.