An Examination of Potential Training Regression Recognition Algorithms for Pilot Training Next

Abstract

The initiative to reduce the Air Forces serious pilot shortage lead to the Pilot Training Next (PTN) program. Under PTN, student pilots progress at an individual rate while making increased use of simulator-based training resources. A previous thesis used data from the first PTN class to conceptualize and prototype a student training flight scheduler. This scheduler did not consider training events required to bring students back to achieved levels of performance if in fact that student performance had regressed. This thesis examines three classes of PTN student data to determine whether student regression in training progression can be detected. A visual and two machine learning-based methods are examined and found to not predict training regression in PTN student pilots.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2021
Accession Number
AD1131027

Entities

People

  • Alex R Gaines

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Air Platforms
  • Autonomy
  • Energy and Power Technologies
  • Weapons Technologies

DTIC Thesaurus Topics

  • Air Force
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Commerce
  • Control Systems
  • Data Mining
  • Dimensionality Reduction
  • Education
  • Flight Simulators
  • Flight Training
  • Individualized Training
  • Information Science
  • Information Systems
  • Literature Surveys
  • Machine Learning
  • Models
  • Neural Networks
  • Psychology
  • Recurrent Neural Networks
  • Simulators
  • Students
  • Training
  • United States Government

Fields of Study

  • Education

Readers

  • Aviation Science / Aeronautics.
  • Instructional Design and Training Evaluation.
  • Regression Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks