Data Analytics For Large Sensor Systems

Abstract

In recent months, researchers at Virginia Tech have performed a pilot study to examine the feasibility of using data analytics to automate the process of assessing sensors and assessing the success of experimental tests. Researchers from the Departments of Statistics and of Aerospace and Ocean Engineering teamed up to apply statistical methods to large sensor count experimental data sets collected in the Virginia Tech Stability Tunnel. This work has been successful in showing that relatively simple methods may be useful in providing rapid feedback to experimentalists about the performance of large-scale measurement systems. The purpose of this document is to propose an expansion of this exploratory effort with the goal of automating the large data sets with measurement and sensor imperfections using statistical methods. These methods would be used to evaluate the quality and content of data in order to guide researchers to increase efficiency and effectiveness of their experiments. In pursuit of this goal, sensor metrics will be identified and implemented in a streaming or rapid calculation system to provide evaluation in either real time or rapid evaluation of quality. The culmination of this work will include a full scale live test of this system as part of a large scale wind tunnel test involving a number of diverse large-scale sensor systems. The work will last 2 years. Total cost for the work proposed is $260,752.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 12, 2016
Source ID
N000141512326

Entities

People

  • Eric P. Smith

Organizations

  • Office of Naval Research
  • United States Navy
  • Virginia Tech

Tags

Readers

  • Organizational Process Management (OPM).
  • Regression Analysis.
  • Research Science/Academic Research

Technology Areas

  • Space