Tableau-Based System for Spatial Reasoning about Directional Relations

The management of qualitative spatial information is an important research area in computer science and AI. Modal logic provides a natural framework for the formalization and implementation of qualitative spatial reasoning. Unfortunately, when directional relations are considered, modal logic systems for spatial reasoning usually turn out to be undecidable (often even not recursively enumerable). In this paper, we give a first example of a decidable modal logic for spatial reasoning with directional relations, called Weak Spatial Propositional Neighborhood Logic (WSpPNL for short). WSpPNL features two modalities, respectively an east modality and a north modality, to deal with non-empty rectangles over N X N. We first show the NEXPTIME-completeness of WSpPNL, then we develop an optimal tableau method for it.