Purpose-Aware Reasoning about Interoperability of Heterogeneous Training Systems

Abstract

Abstract. We describe a novel approach by which software can assess the ability of a confederation of heterogeneous systems to interoperate to achieve a given purpose. The approach uses ontologies and knowledge bases (KBs) to capture the salient characteristics of systems, on the one hand, and of tasks for which these systems will be employed, on the other. Rules are used to represent the conditions under which the capabilities provided by systems can fulfill the capabilities needed to support the roles and interactions that make up each task. An Analyzer component employs these KBs and rules to determine if a given confederation will be adequate, to generate suitable confederations from a collection of available systems, to pre-diagnose potential interoperability problems that might arise, and to suggest system configuration options that will help to make interoperability possible. We have demonstrated the feasibility of this approach using a prototype Analyzer and KBs.

Ontological analysis of terrain data

“…These computed properties of the terrains are compared with the requirements of the particular tasks for which the terrains are intended, using our task analyzer for purposeaware reasoning [2]. For example, the terrain accuracy requirements of a flight simulator training task might be lower than those for an urban ground training task, and a task that needs two terrains may have specific correlation requirements depending on the planned types of interactions between the simulations.…”

Policy-Based Data Downgrading: Toward a Semantic Framework and Automated Tools to Balance Need-to-Protect and Need-to-Share Policies

“…Our PBD approach is based on the ontological framework developed under a U.S. Department of Defense sponsored project called the Open Net-centric Interoperability Standards for Training and Testing (ONISTT) [12], [13], [14] which provides a semantically rich means of describing machine-to-machine interactions in terms of a hierarchical construct known as the Task. The Tasks (referred to as activities in this paper) are described in terms of the capabilities needed to perform that activity (i.e., the data types with their qualitative or quantitative characteristics needed to perform that activity).…”

Reasoning about Resources and Hierarchical Tasks Using OWL and SWRL

“…We define the function H, denoting the Horn clauses of a task, in the following way (keeping the universal quantification of the variables in the clauses implicit). For an abstract task A with refining tasks R 1 …”

“…The translation uses the wellknown correspondence of a large subset of owl, called DLP [7], to Horn clauses (the translation is the same as in our previous work [1]). This means that not all of owl's semantics is covered (i.e., the query answering is not complete), but in practice we have not found this to be a limitation for the ontologies that we work with, as we do not tend to rely on the more complex owl axioms and the inferences that they would enable.…”

“…In [1], we described an approach for automated analysis of military training events. The approach used owl ontologies describing the systems and the purpose for which they were to interoperate.…”