Loading...

Spacecraft Control for Capturing Space Debris via Machine Learning Methods

Alavi Arjas, Mohammad Hassan | 2023

71 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56301 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Kiani, Maryam
  7. Abstract:
  8. The primary purpose of the present research is development and implementation of advanced state estimation and control techniques for space rendezvous and docking. To achieve this aim, the present study has first investigated the use of graph neural networks (GNNs) to filter out the noise of the sensor data in the state estimation process. The measurement package is consisted of a gyroscope, star trackers, and a GPS sensor providing inputs to the GNNs. The obtained results showed that the use of GNNs significantly improves the accuracy of the state estimation compared to traditional methods. In addition, the study has focused on developing advanced control techniques for spacecraft position and attitude control. Q-Learning, Actor Critic Method, and DDPG algorithms have been examined and compared for spacecraft position control and trajectory design. The DDPG algorithm has been selected as the best one and utilized for the simulations in 3D. The obtained simulation results have showen the better performance of the DDPG algorithm than other methods in long-distance trajectory design (about 150 km). The DDGP algorithm has also been adopted for the spacecraft attitude control yielding acceptable performance and accuracy. The goal of the attitude control was to align the robotic arm with space debris for its capture. The study also has designed a robotic arm using the transform2act method, which aims to minimize the control effort of the spacecraft. Finally, the present study has used a simple DeepRL method to stabilize the spacecraft base during the capturing process. In conclusion, this study successfully has developed and implemented learning-based state estimation and control techniques for spacecraft applications using GNNs and advanced control algorithms such as DDPG. The results showed significant improvements in accuracy and performance, making these techniques suitable for future space missions
  9. Keywords:
  10. Reinforcement Learning ; Machine Learning ; Graph Neural Network ; Space Debris ; Spacecraft Guidance ; State Estimation ; Spacecraft Control ; Spacecraft Rendezvous and Docking

 Digital Object List

 Bookmark

No TOC