Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Motion Planning Scheme for Collision Free Autonomous Target Acquisition of Mobile Spacecraft Platform Using Reinforcement Learning
Luleå University of Technology, Department of Computer Science, Electrical and Space Engineering.
2023 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Despite the promising results of reinforcement learning in autonomous maneuvers for ground-based applications, its potential for autonomous target acquisition and collision avoidance in space applications has yet to be thoroughly investigated. Therefore, this study attempts to develop a reinforcement learning-based resilient motion planning approach for autonomous target acquisition and collision avoidance, while focusing on the application where a fying robot is providing assistance to astronauts under microgravity conditions inside the International Space Station. Nevertheless, these complex motion planning strategies require comprehensive validations before deployment on real hardware. Therefore, the study is designed around a ground-based mobile spacecraft platform. The problem at hand has been further simplifed to a 3-Degree of Freedom problem as these types of experiments initially tested and validated in simulated microgravity laboratory environments located on earth. During this research, reinforcement learning agents have been trained to perform tasks, including a) target acquisition in obstacle-free environments, b) target acquisition while avoiding static obstacles, and c) target acquisition while avoiding dynamic obstacles. In all tasks, the trained agents efciently reach the target conditions while successfully avoiding collisions with obstacles.

Place, publisher, year, edition, pages
2023. , p. 42
Keywords [en]
Reinforcement Learning, AI, Space Robotics, International Space Station, Free Flying Robots, Control
National Category
Engineering and Technology Aerospace Engineering
Identifiers
URN: urn:nbn:se:ltu:diva-101378OAI: oai:DiVA.org:ltu-101378DiVA, id: diva2:1798274
Subject / course
Student thesis, at least 30 credits
Educational program
Space Engineering, master's level (120 credits)
Supervisors
Examiners
Available from: 2023-09-19 Created: 2023-09-19 Last updated: 2023-09-19Bibliographically approved

Open Access in DiVA

No full text in DiVA

Search in DiVA

By author/editor
Athauda, A M Bope Gedara Dharshana Ajith
By organisation
Department of Computer Science, Electrical and Space Engineering
Engineering and TechnologyAerospace Engineering

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 182 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf