Deep Reinforcement Learning Applied to Spacecraft Attitude Control and Moment of Inertia Estimation via Recurrent Neural Networks

Abstract

This study investigated two distinct problems related to unknown spacecraft inertia. The first problem explored the use of a recurrent neural network to estimate spacecraft moments of inertia using angular velocity measurements. Initial results showed that, for the configuration examined, the neural network can estimate the moments of inertia when there is a known external torque. The second problem trained a reinforcement learning agent, via proximal policy optimization, to control the attitude of a spacecraft. The results demonstrated that reinforcement learning may be a viable option for guidance and control solutions where the spacecraft model may be unknown. The trained agents displayed a degree of autonomy with their ability to recover from events never experienced in training.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 2021
Accession Number
AD1139779

Entities

People

  • Nathaniel Enders

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Air Force
  • Aircrafts
  • Algorithms
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Control Systems
  • Information Science
  • Machine Learning
  • Model Predictive Control
  • National Security
  • Neural Networks
  • Recurrent Neural Networks
  • Reinforcement Learning
  • Space Force
  • Space Systems
  • Spacecraft
  • Unmanned Aerial Vehicles

Readers

  • Neural Network Machine Learning.
  • Space Exploration and Orbital Mechanics.
  • Tribology (the study of the boundary interaction between sliding surfaces, lubrication, wear and friction).

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Neural Networks
  • Space
  • Space - Spacecraft Maneuvers