OM: an intelligent tutor based on a qualitative model

Abstract This paper describes a causal reasoning system and its application to an intelligent tutoring system for the space domain. The tutoring system was developed initially on a Xerox 1109 using an object-oriented programming language and provides an environment for discovering principles of ground tracks as a direct function of the orbital elements. The knowledge for this discovery environment is coded using a qualitative model of how different shape descriptors can be used to characterize orbit types. This paper looks at some of the practical design and implementation issues associated with causal reasoning systems as embodied in intelligent tutoring systems. It also reports on the results of a study comparing experienced and beginning students' interactions with the intelligent tutoring system.