Deriving Invariants and Constraints from Action Theories

Article type: Research Article

Authors: Zhang, Yan | Foo, Norman Y.

Affiliations: Department of Computing, University of Western Sydney, Nepean, Kingswood, NSW 2747, Australia. E-mail: yan@st.nepean.uws.edu.au | Department of Artificial Intelligence, School of Computer Science and Engineering, University of New South Wales, Sydney, NSW 2052, Australia. E-mail: norman@cse.unsw.edu.au

Abstract: Recent work on reasoning about action has shown that there exists an interesting connection between action specifications and state constraints — it is possible to extract state constraints from action specifications. This work provides us another way to describe the behaviour of dynamic systems. In this paper, we address the problem of generating action invariants from action specifications, and generalizing action invariants into state constraints. We first propose a persistence-based formalism of actions, and show that the generation of action invariants is achieved from action specifications by reasoning about persistence. We then investigate the generalization of action invariants into state constraints. Blocks world examples illustrate the general procedure throughout.

Keywords: knowledge representation, logics for artificial intelligence, reasoning about action

DOI: 10.3233/FI-1997-30108

Journal: Fundamenta Informaticae, vol. 30, no. 1, pp. 109-123, 1997