| The doctoral research is to develop an autonomous intelligent robotic manipulator technology for on-orbit servicing (OOS). More specifically, the research is focused on one of the most critical tasks in OOS- the capture of a non-cooperative object whilst minimizing impact forces and accelerations. The objective of the research is: the development of a vision-based control theory, and the implementation and testing of the developed theory by designing and constructing a custom non-redundant holonomic robotic manipulator. The research validated the newly developed control theory and its ability to (i) capture a moving target autonomously and (ii) minimize unfavourable contact dynamics during the most critical parts of the capture operations between the capture satellite and a non-cooperative/tumbling object. A custom robotic manipulator functional prototype has been designed, assembled, constructed, and programmed from concept to completion in order to provide full customizability and controllability in both the hardware and the software. Based on the test platform, a thorough experimental investigation has been conducted to validate the newly developed control methodologies to govern the behaviour of the robotic manipulators (RM) in an autonomous capture. The capture itself is effected on non-cooperative targets in zero-gravity simulated environment. The RM employs a vision system, force sensors, and encoders in order to sense its environment. The control is effected through position and pseudo-torque inputs to three stepper motors and three servo motors. The controller is a modified hybrid force/neural network impedance controller based on N. Hogan's original work. The experimental results demonstrate the set objectives of this thesis have been successfully achieved. |
