GQR: A Fast Solver for Binary Qualitative Constraint Networks

Matthias Westphal, Stefan Wölfl, Zeno Gantner

Qualitative calculi are constraint-based representation formalisms that allow for efficient reasoning about continuous aspects of the world with inherently infinite domains, such as time and space. GQR (Generic Qualitative Reasoner) is a tool that provides reasoning services for arbitrary binary qualitative calculi. Given qualitative information expressible in a qualitative calculus, GQR checks whether this information is consistent w.r.t. the calculus definition. GQR employs state-of-the-art techniques in both qualitative and constraint rea- soning, such as heuristic search and usage of known tractable subclasses. In contrast to specialized reasoners, it offers reasoning services for a variety of calculi known in the literature, which can be defined in a simple specification language. The main focus in the design and implementation of GQR is to provide a generic and extensible solver that preserves efficiency and scalability.

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.