No description available.

No description available.

No description available.

No description available.

No description available.

A dissertation submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of requirements for the degree of Master of Science. December 14, 2015 / Developing a robot that can operate autonomously is an active area in robotics research. An autonomously operating robot can have a tremendous number of applications such as: surveillance and inspection; search and rescue; and operating in hazardous environments. Reinforcement learning, a branch of machine learning, provides an attractive framework for developing robust control algorithms since it is less demanding in terms of both knowledge and programming effort. Given a reward function, reinforcement learning employs a trial-and-error concept to make an agent learn. It is computationally intractable in practice for an agent to learn “de novo”, thus it is important to provide the learning system with “a priori” knowledge. Such prior knowledge would be in the form of demonstrations performed by the teacher. However, prior knowledge does not necessarily guarantee that the agent will perform well. The performance of the agent usually depends on the reward function, since the reward function describes the formal specification of the control task. However, problems arise with complex reward function that are difficult to specify manually. In order to address these problems, apprenticeship learning via inverse reinforcement learning is used. Apprenticeship learning via inverse reinforcement learning can be used to extract a reward function from the set of demonstrations so that the agent can optimise its performance with respect to that reward function. In this research, a flight controller for the Ar.Drone quadrotor was created using a reinforcement learning algorithm and function approximators with some prior knowledge. The agent was able to perform a manoeuvre that is similar to the one demonstrated by the teacher.

Giang Tsz Sheung Keith. / "Architecture Department, Chinese University of Hong Kong, Master of Architecture Programme 2001-2002, design report." / Includes bibliographical references (leaves 93). / Chapter 00 --- synopsis --- p.Page 01 / Chapter 01 --- Theme and Concept --- p.Page 03 / Chapter 02 --- Research --- p.Page 15 / Chapter 03 --- Site --- p.Page 29 / Chapter 04 --- Program --- p.Page 37 / Chapter 05 --- Master plan --- p.Page 46 / Chapter 06 --- Preliminary design concept / ideas --- p.Page 50 / Chapter 07 --- Design Development --- p.Page 60 / Chapter 08 --- Design / special studies --- p.Page 78 / Chapter 09 --- Design / general public experience --- p.Page 83 / Chapter 10 --- Design / cadet pilot experience --- p.Page 87 / Chapter 11 --- Schedule --- p.Page 92 / Chapter 12 --- Reference --- p.Page 93 / Chapter 13 --- Appendix --- p.Page 94 / Chapter 14 --- Acknowledgement --- p.Page 98

abstract: This study was undertaken to ascertain to what degree, if any, virtual reality training was superior to monitor based training. By analyzing the results in a 2x3 ANOVA it was found that little difference in training resulted from using virtual reality or monitor interaction to facilitate training. The data did suggest that training involving rich textured environments might be more beneficial under virtual reality conditions, however nothing significant was found in the analysis. It might be possible that significance could be obtained by comparing a virtual reality set-up with higher fidelity to a monitor trial. / Dissertation/Thesis / Masters Thesis Engineering 2019

Thesis (Ph. D.)--University of Washington, 2006. / Vita. Includes bibliographical references (leaves 168-172).

The purpose of this study was to evaluate the suitability of using an existing 25-ft radius centrifuge as a dynamic flight simulator for “full mission” F/A-18 strike fighter mission training with respect to the representativeness of pilot-perceived motion and acceleration cues. The methodology employed in this study consisted of analyzing F/A-18 mission tasks, collecting pilot opinion surveys of important sensory cues needed in simulator training, and conducting an analysis of human pilot perceptual problems caused by centrifuge motion constraints. This study identified a number of issues indicating that a centrifuge-based flight simulator shows limited potential for use in “full mission” F/A-18 training scenarios. Specifically, there is a fundamental mismatch between the 6 degree-of-freedom mission-representative acceleration environment experienced in the aircraft and the 3 degree-of-freedom acceleration environment the centrifuge is able to provide. The centrifuge is not optimized for the typical acceleration environment experienced during F/A-18 missions and has significant limitations in “near one g” and “near zero g” flight conditions. Additionally, the centrifuge causes a variety of undesired, unrealistic, and debilitating vestibular artifacts that are not consistent with what a pilot experiences in the aircraft when performing the same mission task, degrading the effectiveness of training. Despite its limited suitability as a “full mission” F/A-18 simulator, the centrifuge is an essential physiological training device, shows good potential as a part-task trainer for departure/spin training, and should continue to play a role in the F/A-18 training continuum.