Visual Servo Based Space Robotic Docking for Active Space Debris Removal

This thesis developed a 6DOF pose detection algorithm using machine learning capable of providing the orientation and location of an object in various lighting conditions and at different angles, for the purposes of space robotic rendezvous and docking control. The computer vision algorithm was paired with a virtual robotic simulation to test the feasibility of using the proposed algorithm for visual servo. This thesis also developed a method for generating virtual training images and corresponding ground truth data including both location and orientation information. Traditional computer vision techniques struggle to determine the 6DOF pose of an object when certain colors or edges are not found, therefore training a network is an optimal choice. The 6DOF pose detection algorithm was implemented on MATLAB and Python. The robotic simulation was implemented on Simulink and ROS Gazebo. Finally, the generation of training data was done with Python and Blender.

Author dc:creator

• Lal, Sukhjinder Singh

Advisor dc:contributor.advisor

• Zhu, George Z.H.

Statement dc:rights

• Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.

Language dc:language

en

Handle dc:identifier.uri

http://hdl.handle.net/10315/39128

OAI identifier oai:identifier

oai:yorkspace.library.yorku.ca:10315/39128