A Novel Semi-Supervised Kernel Formulation for Extrapolation from Small Datasets: Rapid Predictive Modeling of the Effect of a Leeway Object Geometry on its Drift and Divergence in Deep Waters

Abstract

This project creates novel mathematical models and innovative algorithms to 1. reduce the number of required tests, 2. Increase the amount of information extracted from test results, and 3. enhance the transfer of knowledge between test scenarios, to vertically advance the science of test and evaluation. The resulting models and algorithms are used to leverage expensive hard-to-get test data from deep water to tackle the complex problem of dynamic prediction of the drift and divergence of leeway objects based on the geometry and air and surface forces.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 21, 2022
Source ID
FA95502110417XX0

Entities

People

  • Adel Alaeddini

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Texas at San Antonio

Tags

Readers

  • Combustion and Flow Dynamics.
  • Geodesy
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks