Neuromorphic Computing for Very Large Test and Evaluation Data Analysis
Abstract
The research conducted in this project can be broken down into two parallel tracks, development of self-reconfigurable neuromorphic computing architectures utilizing memristive nanotechnology, and analysis and utilization of newly available hardware-based artificial neural network chips. These two aspects of the program were complementary. The neuromorphic architectures research focused on long term disruptive technologies with high risk but revolutionary potential. The hardware-based neural network research provided the means to apply many of the advantages of biologically inspired processing using advanced chipsets that are available today. Overall, hardware-based neural processing research allowed us to study the fundamental system and architectural issues relevant for employing neuromorphic computing concepts towards the development of autonomous systems for perception and data analysis. The utilization of computationally intelligent processors for tasks within the United States Air Force Developmental Test and Evaluation (T&E) community was emphasized, with a focus on test data analysis and process control efficiencies. This program resulted in the successful implementation of reconfigurable neuromorphic architecture building blocks. In addition, this work demonstrated the benefits of neuromorphic architectures for increased performance over conventional computing in certain applications, especially in Size, Weight and Power (SWaP) constrained environments.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 01, 2014
- Accession Number
- ADA603822
Entities
People
- Bryant Wysocki
- Clare Thiem
- James Bohl
- Nathan D. McDonald
- Thomas Renz
Organizations
- Air Force Research Laboratory