Polynomial Networks for Testing Flight Critical Systems

The objective of this effort is to demonstrate a prototype model of a flight-critical system using AbTech's AIM network synthesis tool and NASA's CLIPS production rule system and to demonstrate the ability of an AIM generated polynomial network to reliably perform sensitivity analysis using variable components of this flight critical system. The resulting adaptive test generation, analysis, and control modeling system will demonstrate the feasibility of developing a revolutionary test definition and analysis product capable of adapting in a dynamic environment, monitoring and diagnosing complex system behavior, and predicting future faults and maintenance requirements. The innovation is the application of learning software to modeling, sensitivity analysis, and test- synthesis and diagnosis. The AIM test system is designed to enhance a systems engineer's ability to understand system interactions, make decisions on a systems ability to complete a desired function, determine the effect that a component has on the overall system, and perform system validation. In Phase I we will demonstrate the ability of AIM to model a flight critical system, perform sensitivity analysis, and to test for validating functionality and operability.