Biologically Inspired Automatic Target Detection, Classification, and Tracking
Abstract
The decision to use and deliver kinetic and/or non-kinetic fires has been and will forever been twined into war. The U.S. military and specifically the Marines have been the masters of this process but to maintain this superiority, fire support needs to become more timely, discriminatory, lethal, and effective intoday's data-saturated environment. Also, with the distributive nature of the modern operating environment, producing a SWaP-T-compatible solution is vital. To bridge these gaps, the author proposes to offload the target detection, classification, and tracking to a biologically inspired automated system composed of a Dynamic Vision Sensor and a spiking neural network running on neuromorphic hardware. Emphasis was placed on the spiking neural network algorithm development and building/evaluating the system. The author found that this approach could yield a system that will provide the warfighter with actionable information to improve the kill chain process while minimizing power consumption and time taken at the point of collection. The hope is that the research presented here will spur advances in the field of biologically inspired neuromorphic platforms that will produce timely, accurate, distilled, and actionable information to the end user to offload mundane/trivial tasks to allow for more decision time and space.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2020
- Accession Number
- AD1126477
Entities
People
- Eric H. Kim
Organizations
- Naval Postgraduate School