Thesis (M. Phil.)--University of Hong Kong, 1998. / Includes bibliographical references (leaves 41-42).

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Vita. Includes bibliographical references (leaves 271-279). Available also in a digital version from Dissertation Abstracts.

Thesis (M.S.)--West Virginia University, 2005. / Title from document title page. Document formatted into pages; contains vii, 88 p. : ill. Includes abstract. Includes bibliographical references (p. 83-88).

Thesis (M.S.)--West Virginia University, 2005. / Title from document title page. Document formatted into pages; contains viii, 104 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 89-92).

Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2006. / Includes bibliographical references (p. 31-33).

by Siu Wai-man, Raymond. / Bibliography: leaves 75-76 / Thesis (M.B.A.)--Chinese University of Hong Kong, 1986

So Wai Kuen. / Year shown on spine: 1997. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (leaves 87-92). / INTRODUCTION --- p.1 / Chapter 1.1 --- Introduction to SLCR Problem --- p.1 / Chapter 1.2 --- Our Contributions --- p.1 / Chapter 1.3 --- Organization of the thesis --- p.3 / LITERATURE REVIEW --- p.4 / Chapter 2.1 --- Research in the Capacitated Resource Constraint Problem --- p.4 / Chapter 2.2 --- The Single Level Constrained Resource Problem --- p.5 / Chapter 2.3 --- The Multiple Level Constrained Resource Problem --- p.8 / Chapter 2.4 --- Research in the Fixed Charge Problem --- p.9 / Chapter 2.4.1 --- Approximate Methods --- p.9 / Chapter 2.4.2 --- Exact Methods --- p.10 / Chapter 2.5 --- Conclusion --- p.11 / THE SLCR PROBLEM WITH BACKORDERING --- p.12 / Chapter 3.1 --- Problem Description and Formulation --- p.13 / Chapter 3.2 --- Description of the heuristic --- p.19 / Chapter 3.2.1 --- Phase I --- p.19 / Chapter 3.2.2 --- Phase II --- p.26 / Chapter 3.3 --- Design of Computational Experiments --- p.30 / Chapter 3.3.1 --- Specifications of test problems ( 3 products and 12 period case ) --- p.31 / Chapter 3.3.2 --- Computation of the Lower Bound --- p.38 / Chapter 3.4 --- Computational Results --- p.39 / Chapter 3.5 --- Comparison to Millar and Yang's Algorithm --- p.48 / Chapter 3.5.1 --- Comparison Results --- p.49 / Chapter 3.6 --- Conclusion --- p.50 / THE OPTIMIZATION BASED ALGORITHM --- p.51 / Chapter 4.1 --- The Formulation --- p.52 / Chapter 4.2 --- The Algorithm --- p.60 / Chapter 4.2.1 --- Phase I --- p.60 / Chapter 4.2.2 --- Phase II --- p.63 / Chapter 4.2.3 --- Phase III --- p.70 / Chapter 4.3 --- An Illustrative Example --- p.72 / Chapter 4.4 --- Computational Results --- p.79 / Chapter 4.5 --- Conclusion --- p.84 / CONCLUSION --- p.85 / BIBLIOGRAPHY --- p.87

This thesis represents a relatively new aspect of computing with regard to robotics. The need for fast, efficient collision avoidance algorithms is growing rapidly. Because conventional methods are complex and require vast amounts of computation, heuristic algorithms are more appealing. The focus of this thesis is the problem of moving a point through three dimensional space while avoiding known polyhedral obstacles. A heuristic algorithm to find shortest (near-optimal) collision-free paths in the presence of polyhedral obstacles, given initial and final positions, is presented. Previous methods for the problem rely on an a priori discretization of the space. The points in the discretization form nodes of a graph, and the collision avoidance problem is then solved by using some shortest path algorithm on the graph. The heuristic suggested here successively adds nodes to a graph, thus keeping the size of the graph manageable. The computational results are extremely encouraging.

Providing robots with the ability to grasp objects has, despite decades of research, remained a challenging problem. The problem is approachable in constrained environments where there is ample prior knowledge of the scene and objects that will be manipulated. The challenge is in building systems that scale beyond specific situational instances and gracefully operate in novel conditions. In the past, heuristic and simple rule based strategies were used to accomplish tasks such as scene segmentation or reasoning about occlusion. These heuristic strategies work in constrained environments where a roboticist can make simplifying assumptions about everything from the geometries of the objects to be interacted with, level of clutter, camera position, lighting, and a myriad of other relevant variables. With these assumptions in place, it becomes tractable for a roboticist to hardcode desired behaviour and build a robotic system capable of completing repetitive tasks. These hardcoded behaviours will quickly fail if the assumptions about the environment are invalidated. In this thesis we will demonstrate how a robust grasping system can be built that is capable of operating under a more variable set of conditions without requiring significant engineering of behavior by a roboticist. This robustness is enabled by a new found ability to empower novel machine learning techniques with massive amounts of synthetic training data. The ability of simulators to create realistic sensory data enables the generation of massive corpora of labeled training data for various grasping related tasks. The use of simulation allows for the creation of a wide variety of environments and experiences exposing the robotic system to a large number of scenarios before ever operating in the real world. This thesis demonstrates that it is now possible to build systems that work in the real world trained using deep learning on synthetic data. The sheer volume of data that can be produced via simulation enables the use of powerful deep learning techniques whose performance scales with the amount of data available. This thesis will explore how deep learning and other techniques can be used to encode these massive datasets for efficient runtime use. The ability to train and test on synthetic data allows for quick iterative development of new perception, planning and grasp execution algorithms that work in a large number of environments. Creative applications of machine learning and massive synthetic datasets are allowing robotic systems to learn skills, and move beyond repetitive hardcoded tasks.

No description available.