Advanced diagnostic/prognostic reasoning and evidence transformation techniques for improved avionics maintenance

Abstract

Techniques used in the design and implementation of modern avionics suggest an opportunity to re-examine the maintenance and repair process for current and future systems. The authors have developed a framework with automated evidence collection, data representation and storage, and, advanced automated reasoning techniques to implement within an avionics health management system. This paradigm shift approach utilizes advanced capture and representation of environmental, operational, and component inter-relationship evidence at a systems level to reduce diagnostic ambiguity, guide at-wing testing, and provide prognostics. Modular and reusable software and data elements that combine built-in-test (BIT) with contextual information, component usage models, and evidentiary reasoning techniques are described. A use case scenario is presented that illustrates evidence collection and XML transformation, automated database storage and retrieval, and automated advanced reasoning for diagnostics and prognostics, and finally, an example of at wing ambiguity reduction using an advanced evidence-based reasoner is presented.