After capturing the object, the space robot must complete the following two tasks: one is to berth the object, and the other is to re-orientate the attitude of the whole robot system for communication and power supply. Therefore, I propose a method to accomplish these two tasks simultaneously using manipulator motion only. / Finally I propose a novel approach based on Genetic Algorithms (GAs) to optimize the approach trajectory of space robots in order to realize effective and stable operations. I complete the minimum-torque path planning in order to save the limited energy in space, and design the minimum jerk trajectory for the stabilization of the space manipulator and its space base. These optimal algorithms are very important and useful for the application of space robot. / In this thesis, I study and analyze the dynamics and control problems of space robot for capturing objects. This work has potential impact in space robotic applications. I first study the contact and impact dynamics of space robot and objects. I specifically focus on analyzing the impact dynamics and mapping the relationship of influence and speed. Then, I develop the fundamental theory for planning the minimum-collision based trajectory of space robot and designing the configuration of space robot at the moment of capture. / Space robots are expected to perform intricate tasks in future space services, such as satellite maintenance, refueling, and replacing the orbital replacement unit (ORU). To realize these missions, the capturing operation may not be avoided. Such operations will encounter some challenges because space robots have some unique characteristics unfound on ground-based robots, such as, dynamic singularities, dynamic coupling between manipulator and space base, limited energy supply and working without a fixed base, and so on. In addition, since contacts and impacts may not be avoided during capturing operation. Therefore, dynamics and control problems of space robot for capturing objects are significant research topics if the robots are to be deployed for the space services. A typical servicing operation mainly includes three phases: capturing the object, berthing and docking the object, then repairing the target. Therefore, this thesis will focus on resolving some challenging problems during capturing the object, berthing and docking, and so on. / The ultimate goal of space services is to realize the capture and manipulation autonomously. Therefore, I propose an affective approach based on learning human skill to track and capture the objects automatically in space. With human-teaching demonstration, the space robot is able to learn and abstract human tracking and capturing skill using an efficient neural-network learning architecture that combines flexible Cascade Neural Networks with Node Decoupled Extended Kalman Filtering (CNN-NDEKF). The simulation results attest that this approach is useful and feasible in tracking trajectory planning and capturing of space robot. / To compensate for the attitude of the space base during the capturing approach operation, a new balance control concept which can effectively balance the attitude of the space base using the dynamic couplings is developed. The developed balance control concept helps to understand of the nature of space dynamic coupling, and can be readily applied to compensate or minimize the disturbance to the space base. / Huang Panfeng. / "December 2005." / Adviser: Yang Sheng Xu. / Source: Dissertation Abstracts International, Volume: 67-11, Section: B, page: 6693. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 133-143). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

published_or_final_version / Mechanical Engineering / Master / Master of Philosophy

Many important real-world robotic tasks have high diameter, that is, their solution requires a large number of primitive actions by the robot. For example, they may require navigating to distant locations using primitive motor control commands. In addition, modern robots are endowed with rich, high-dimensional sensory systems, providing measurements of a continuous environment. Reinforcement learning (RL) has shown promise as a method for automatic learning of robot behavior, but current methods work best on lowdiameter, low-dimensional tasks. Because of this problem, the success of RL on real-world tasks still depends on human analysis of the robot, environment, and task to provide a useful set of perceptual features and an appropriate decomposition of the task into subtasks. This thesis presents Self-Organizing Distinctive-state Abstraction (SODA) as a solution to this problem. Using SODA a robot with little prior knowledge of its sensorimotor system, environment, and task can automatically reduce the effective diameter of its tasks. First it uses a self-organizing feature map to learn higher level perceptual features while exploring using primitive, local actions. Then, using the learned features as input, it learns a set of high-level actions that carry the robot between perceptually distinctive states in the environment. Experiments in two robot navigation environments demonstrate that SODA learns useful features and high-level actions, that using these new actions dramatically speeds up learning for high-diameter navigation tasks, and that the method scales to large (buildingsized) robot environments. These experiments demonstrate SODAs effectiveness as a generic learning agent for mobile robot navigation, pointing the way toward developmental robots that learn to understand themselves and their environments through experience in the world, reducing the need for human engineering for each new robotic application. / text

published_or_final_version / abstract / toc / Industrial and Manufacturing Systems Engineering / Master / Master of Philosophy

No description available.

No description available.

No description available.

No description available.

No description available.

The research developed force-motion strategies and subsequent force and position control algorithms, using a PUMA 560 robot arm and its original controller. A task decomposition methodology has been developed that enables a mechanical assembly task to be subdivided into a series of executable subtasks. By applying this methodology to the assembly of a hydraulic gear pump, a library of special purpose, task oriented, subtask programs were created. Most of these programs, though derived for a pump assembly task, are applicable (when used with appropriate parameters) to other assembly tasks. Most of the algorithms require force/torque sensory information that is supplied by a JR³ wrist force sensor. The force control algorithms use that data and system compliance in order to produce new position instructions that are transferred to the controller of the arm. The logic of the control law and system behaviour when contacting the environment, were checked, using the dynamics and compliance of a simplified structure of a robotic arm and its wrist sensor. A demonstration of the pump assembly task, using the arm, force sensor, controller and the derived library algorithms is an integral part of the thesis. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate