Utilizing Neural Networks to Interpret Data Acquired from Automated Test Systems

Patrick J. Sineebaugh, William H. Green

Innovative diagnostic testing techniques must be developed and applied in order to meet the increasing challenges associated with testing complex systems in an era of budget and personnel reductions. Research in testing and evaluation systems in manufacturing at the U.S. Army Research Laboratory Materials Directorate has focused on automating conventional test systems via the development of Intelligent Testing Systems (ITS). An ITS can be defined as a computer based system that utilizes state-of-the-art classification or decision making technology, often artificial intelligence (AI) techniques, to enable the system make decisions or perform functions previously made by human operators. This paper begins by discussing the defining characteristics and advantages of automated test systems. This is followed by a discussion of the advantages of applying neural networks to data pattern analysis and classification. The reasons for using the backlm3pagation neural network algorithm in the case study A Smart Shock Absorber Test Stand (SSATS) are then given. The motivation for and the development of the SSATS system is described. Fmally, this paper describes the benefits of utilizing the SSATS system and of implementing the methods used to develop it to other Intelligent Testing Systems.

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.