(Some) Default and Non-Monotonic Aspects of Qualitative Spatial Reasoning: A Preliminary Report

Mehul Bhatt

We present some scenarios where default and/or non-monotonic reasoning patterns are either necessary or useful for the modelling of dynamic spatial environments. The identified instances bear a direct relationship to the fundamental epistemological issues relevant to the frame and ramification problems; these are cases where a typical use of non-monotonicity is necessary at a meta-theoretical or domain-independent level. Furthermore, non-monotonic reasoning is also necessary whilst modelling the appearance and disappearance of spatial objects -- these phenomena, considered inherent to dynamic spatial systems, essentially involve the representation of an incompletely known domain of discourse. The final case, from the viewpoint of this paper, involves the use of non-monotonic reasoning for modelling causal explanation tasks in an abductive manner. Indeed, the non-monotonic patterns we illustrate are only indicative of the ones encountered and accounted for in the context of our key task of developing a situation calculus based domain-independent qualitative spatial theory that is usable in diverse dynamic spatial domains. The identification of other similar patterns and the general utility of non-monotonic reasoning from a specific spatial reasoning viewpoint is an important research agenda, and this paper calls for a further investigation of the same within the mainstream qualitative spatial reasoning domain.

Subjects: 3.5 Qualitative Reasoning; 3.3 Nonmonotonic Reasoning

Submitted: Apr 28, 2008


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.