The U-curves of five different test objects, three of which contain a rod floating object at different positions within the gap, are characterised. During the testing, a high speed camera was used to photograph the discharges. The results are compared to Rizk’s theoretical model for determining the flashover voltage of gaps with floating objects are presented. It is concluded that the position of the floating object within the gap affects which discharge mechanism exists in each of the gaps. The effect that each discharge mechanism has on the flashover voltage and time to crest of the gap is shown. Time interval photographs are presented showing the formation of a discharge channel due to the streamer mechanism. In evaluating the high speed photographs it is seen that the extent of the branching of the discharge channel is a function of the time to crest of the applied impulse, more branching is evident for shorter times to crest.

Learning object repositories, a form of digital libraries, are robust systems that provide educators new ways to search for educational resources, collaborate with peers, and provide instruction to students in unique and varied ways. This study examines a learning object repository created by a large suburban school district to increase teaching information and encourage collaboration among teachers. Despite investing nearly $2 million to develop the software and seed the repository with materials, data suggest that teacher use falls below set goals. This document explores five years of site traffic, user engagement, social interaction, asset growth, as well as the authoring of instructional materials as a means to evaluate the repository. The results of the study may inform the policy decisions of educational organizations when considering digital learning environments.

Handing over objects is the foundation of many human-robot interaction and collaboration tasks. In the scenario where a human is handing over an object to a robot, the human chooses where the object needs to be transferred. The robot needs to accurately predict this point of transfer to reach out proactively, instead of waiting for the final position to be presented. We first conduct a human-to-robot handover motion study to analyze the effect of user height, arm length, position, orientation and robot gaze on the object transfer point. Our study presents new observations on the effect of robot's gaze on the point of object transfer. Next, we present an efficient method for predicting the Object Transfer Point (OTP), which synthesizes (1) an offline OTP calculated based on human preferences observed in the human-robot motion study with (2) a dynamic OTP predicted based on the observed human motion. Our proposed OTP predictor is implemented on a humanoid nursing robot and experimentally validated in human-robot handover tasks. Compared to using only static or dynamic OTP estimators, it has better accuracy at the earlier phase of handover (up to 45% of the handover motion) and can render fluent handovers with a reach-to-grasp response time (about 3.1 secs) close to natural human receiver's response. In addition, the OTP prediction accuracy is maintained across the robot's visible workspace by utilizing a user-adaptive reference frame.

Yu-shan Chan. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1995. / Includes bibliographical references (leaves 95-99). / Chapter 1. --- INTRODUCTION --- p.1 / Chapter 1.1 --- "Expert System, Expert System Shell, and ""MetaSystem""" --- p.2 / Chapter 1.2 --- Adopting OBJECTBASE In EXPERT SYSTEM SHELL(ESS) --- p.4 / Chapter 2. --- SURVEY ON EXISTING SYSTEMS --- p.7 / Chapter 2.1 --- Review of inference models --- p.7 / Chapter 2.1.1 --- The Classical Period --- p.9 / Chapter 2.1.2 --- The modern period --- p.11 / Chapter 2.2 --- Rules in Objectbase vs. other Representations --- p.12 / Chapter 2.2.1 --- Rule-based systems --- p.13 / Chapter 2.2.2 --- Object-oriented systems --- p.13 / Chapter 2.2.3 --- Other systems --- p.13 / Chapter 2.2.4 --- Rules embedded in object-- the Objectbase approach --- p.14 / Chapter 2.3 --- Conclusion --- p.15 / Chapter 3. --- DESIGN OF ESS FOR AN OBJECTBASE SYSTEM --- p.16 / Chapter 3.1 --- Introducing ESS in Objectbase --- p.18 / Chapter 3.1.1 --- The Concept of Object Modeling --- p.19 / Chapter 3.1.2 --- Why Objectbase? --- p.20 / Chapter 3.1.3 --- ESS : a higher layer on Objectbase --- p.22 / Chapter 3.1.4 --- Schema Objects and Shell Objects --- p.23 / Chapter 3.2 --- Module design of ESS --- p.24 / Chapter 3.2.1 --- Knowledge Representation Module --- p.25 / Chapter 3.2.2 --- Objectbase inference module --- p.27 / Chapter 3.2.3 --- The Rule一Inference Module --- p.28 / Chapter 3.3 --- Knowledge Representation --- p.29 / Chapter 3.3.1 --- Schema Knowledge & the Rulebase --- p.30 / Chapter 3.3.2 --- Rule Structure --- p.31 / Chapter 3.4 --- Inference Engine --- p.35 / Chapter 3.4.1 --- The Two Levels of Inference --- p.35 / Chapter 3.5 --- Rule一Inference (RI) --- p.37 / Chapter 3.5.1 --- Structural design of RI --- p.38 / Chapter 3.5.2 --- Drawing Inference --- p.39 / Chapter 3.5.3 --- Query Processor and RI --- p.42 / Chapter 3.5.4 --- RI and the Inference Engine(IE) --- p.43 / Chapter 3.6 --- Conclusion --- p.43 / Chapter 4. --- IMPLEMENTATION --- p.45 / Chapter 4.1 --- Rulelnference: a comprehensive structure --- p.46 / Chapter 4.1.1 --- Class Rule --- p.46 / Chapter 4.1.2 --- Class RuleList --- p.47 / Chapter 4.1.3 --- Accompany data structures for inference --- p.48 / Chapter 4.1.4 --- Class Rulelnference --- p.49 / Chapter 4.2 --- Rule Setting --- p.51 / Chapter 4.2.1 --- Rule Construction --- p.51 / Chapter 4.2.2 --- Rule Parsing and the Rule Definition Language (RDL) --- p.52 / Chapter 4.3 --- How Inference is done in ESS --- p.53 / Chapter 4.3.1 --- Reset and Load system --- p.53 / Chapter 4.3.2 --- Inference making --- p.54 / Chapter 4.4 --- Using RuleInference in the Rule Constructor --- p.58 / Chapter 4.4.1 --- The Rule Constructor --- p.59 / Chapter 4.5 --- Using Rulelnference in the Application Constructor --- p.60 / Chapter 4.5.1 --- The RiNode --- p.61 / Chapter 4.5.2 --- Schema and Rule Set Handling --- p.63 / Chapter 4.6 --- Conclusion --- p.64 / Chapter 5. --- CASE STUDY --- p.66 / Chapter 5.1 --- Background on Statement analysis --- p.66 / Chapter 5.1.1 --- Ratios for decision making --- p.68 / Chapter 5.2 --- Sample System: Financial Data Analysis System --- p.70 / Chapter 5.2.1 --- The FINANCE schema --- p.71 / Chapter 5.2.2 --- Rules --- p.73 / Chapter 5.2.3 --- Results --- p.75 / Chapter 5.3 --- Evaluation --- p.81 / Chapter 5.4 --- Conclusion --- p.82 / Chapter 6. --- RESULT AND DISCUSSION --- p.84 / Chapter 6.1 --- An Expert System Shell on Objectbase --- p.84 / Chapter 6.2 --- The ESS on MOBILE --- p.85 / Chapter 6.3 --- Pros and cons for the ESS --- p.86 / Chapter 6.4 --- MOBILE: how it has been improved --- p.87 / Chapter 7. --- CONCLUSION --- p.89 / Chapter 7.1 --- Comparison --- p.91 / Chapter 7.2 --- Appraisal --- p.92 / Chapter 8. --- REFERENCES --- p.95 / Table of Content for Appendixes / APPENDIX 1. RULE DEFINITION LANGUAGE --- p.100 / APPENDIX 2. THE CLASS RULEINFERENCE --- p.103 / APPENDIX 3. THE RINODE --- p.104 / APPENDIX 4. FINANCIAL STATEMENT ANALYSIS --- p.108 / APPENDIX 5. DATA STRUCTURE OF RULE AND RULELIST --- p.117 / APPENDIX 6. DATA STRUCTURE OF VARLIST AND ACTLIST --- p.118 / APPENDIX 7. DATA STRUCTURE OF RULEINFERENCE --- p.121

Caching is a widely used technique to improve the scalability of distributed systems. A central issue with caching is maintaining object replicas consistent with their master copies. Large distributed systems, such as the Web, typically deploy heuristic-based consistency mechanisms, which increase delay and place extra load on the servers, while not providing guarantees that cached copies served to clients are up-to-date. Server-driven invalidation has been proposed as an approach to strong cache consistency, but it requires servers to keep track of which objects are cached by which clients. We propose an alternative approach to strong cache consistency, called MONARCH, which does not require servers to maintain per-client state. Our approach builds on a few key observations. Large and popular sites, which attract the majority of the traffic, construct their pages from distinct components with various characteristics. Components may have different content types, change characteristics, and semantics. These components are merged together to produce a monolithic page, and the information about their uniqueness is lost. In our view, pages should serve as containers holding distinct objects with heterogeneous type and change characteristics while preserving the boundaries between these objects. Servers compile object characteristics and information about relationships between containers and embedded objects into explicit object management commands. Servers piggyback these commands onto existing request/response traffic so that client caches can use these commands to make object management decisions. The use of explicit content control commands is a deterministic, rather than heuristic, object management mechanism that gives content providers more control over their content. The deterministic object management with strong cache consistency offered by MONARCH allows content providers to make more of their content cacheable. Furthermore, MONARCH enables content providers to expose internal structure of their pages to clients. We evaluated MONARCH using simulations with content collected from real Web sites. The results show that MONARCH provides strong cache consistency for all objects, even for unpredictably changing ones, and incurs smaller byte and message overhead than heuristic policies. The results also show that as the request arrival rate or the number of clients increases, the amount of server state maintained by MONARCH remains the same while the amount of server state incurred by server invalidation mechanisms grows.

Inspired by nature, swarm robotics aims to increase system robustness while utilizing simple agents. In this work, we present a novel approach to achieve decentralized coordination of forces during collective manipulation tasks resulting in a highly scalable, versatile, and robust solution. In this approach, each robot involved in the collective object manipulation task relies on the behavior of a cooperative ``virtual teammate' in a fully decentralized architecture, regardless of the size and configuration of the real team. By regulating their actions with their corresponding virtual counterparts, robots achieve continuous pose control of the manipulated object, while eliminating the need for inter-agent communication or a leader-follower architecture. To experimentally study the scalability, versatility, and robustness of the proposed collective object manipulation algorithm, a new swarm agent, Δρ is introduced which is able to apply linear forces in any planar direction. Efficiency and effectiveness of the proposed decentralized algorithm are investigated by quantitative performance metrics of settling time, steady-state error, path efficiency, and object velocity profiles in comparison with a force-optimal centralized version that requires complete information. Employing impedance control during manipulation of an object provides a mean to control its dynamic interactions with the environment. The proposed decentralized algorithm is extended to achieve a desired multi-dimensional impedance behavior of the object during a collective manipulation without inter-agent communication. The proposed algorithm extension is built upon the concept of ``virtual coordination' which demands every agent to locally coordinate with one virtual teammate. Since the real population of the team is unknown to the agents, the resultant force applied to the manipulated object would be directly scaled with the team population. Although this scaling effect proves useful during position control of the object, it leads to a deviation from the desired dynamic response when employed in an impedance control scheme. To minimize such deviations, a gradient descent algorithm is implemented to determine a scaling parameter defined on the control action. The simulation results of a multi-robot system with different populations and formations verify the effectiveness of the proposed method in both generating the desired impedance response and estimating the population of the group. Eventually, as two case studies, the introduced algorithm is used in robotic collective manipulation and human- assistance scenarios. Simulation and experimental results indicate that the proposed decentralized communication- free algorithm successfully performs collective manipulation in all tested scenarios, and matches the performance of the centralized controller for increasing number of agents, demonstrating its utility in communication- limited systems, remote environments, and access-limited objects.

Human-swarm interaction is an important milestone for the introduction of swarm-intelligence based solutions into real application scenarios. One of the main hurdles towards this goal is the creation of suitable interfaces for humans to convey the correct intent to multiple robots. As the size of the swarm increases, the complexity of dealing with explicit commands for individual robots becomes intractable. This brings a great challenge for the developer or the operator to drive robots to finish even the most basic tasks. In our work, we consider a different approach that humans specify only the desired goal rather than issuing individual commands necessary to obtain this task. We explore this approach in a collaborative transport scenario, where the user chooses the target position of an object, and a group of robots moves it by adapting themselves to the environment. The main outcome of this thesis is the design of integration of a collaborative transport behavior of swarm robots and an augmented reality human interface. We implemented an augmented reality (AR) application in which a virtual object is displayed overlapped on a detected target object. Users can manipulate the virtual object to generate the goal configuration for the object. The designed centralized controller translate the goal position to the robots and synchronize the state transitions. The whole system is tested on Khepera IV robots through the integration of Vicon system and ARGoS simulator.

Various types of small mobile objects such as recreational unmanned vehicles have become easily approachable devices to the public because of technology advancements. The technology advancements make it possible to manufacture small, light, and easy to control unmanned vehicles, therefore the public are able to handily access those unmanned vehicles. As the accessibility to unmanned vehicles for recreational purposes, accidents or attacks to threat a person using those the unmanned vehicles have been arising and growing rapidly. A specific person could be a target of a threat using an unmanned vehicle in open public places due to its small volume and mobility. Even though an unmanned vehicle approaches to a person, it could be difficult to detect the unmanned vehicle before the person encounters because of the compact size and maneuverability. <div><br></div><div>This research is to develop a radar system that is able to operate in open public areas to detect and track unmanned vehicles. It is not capable using existing radar systems such as for navigation, aviation, national defense, air traffic control, or weather forecasting to monitor and scan public places because of large volume, high operation cost, and danger to human health of the radar systems. For example, if electromagnetic fields emitted from high-power radar penetrate exposed skin surface or eyes, the energy from the electromagnetic fields can cause skin burns, eye cataracts, or more (Zamanian & Hardiman, 2005). Therefore, a radar system that can perform at the public place is necessary for monitoring and surveillance the area. <div><br></div><div>The hardware of this proposed radar system is composed of three parts: 1) radio frequency transmission and receiver part which we will call RF part; 2) transmitting radio frequency control and amplifying reflected signal part which we will call electric part; and 3) data collection, data processing, and visualization part which we will call post-processing part. A transmitting radio frequency control and an amplifying reflected signal part are based on a research performed at a lecture and labs designed by researchers at Massachusetts Institute of Technology (MIT) Lincoln Lab, Charvat et al. (2012) and another lecture and labs designed by a professor at University of California at Davis, Liu (2013). The radar system designed at University of California at Davis is based on the system designed at MIT Lincoln Lab that proposed a design of a small, low cost, and low power consuming radar. The low power radar proposed by MIT Lincoln Lab is suitable to operate in any public places without any restrictions for human health because of it low power transmission, however surveillance area is relatively short and limited. To expand monitoring area with this proposed low power radar system, the transmit power of the radar system proposed in this study is enhanced comparing to the radar proposed by MIT Lincoln Lab. Additionally, the radar system is designed and fabricated on printed circuit boards (PCBs) to make the system compact and easy to access for use of various purposed of research fields. For instance, the radar system can be utilized for mapping, localization, or imaging. <div><br></div><div>The first part of post-processing is data collection. The raw data received and amplified through the electric part in the hardware is collected through a compact computer, a Raspberry Pi 3, that is directly connected to the radar. The data collected every second and the collected data is transferred to the post-processing devices, which is a laptop computer in this research. The post-processing device processes data to estimate range of the object, applies filters for tracking, and visualizes the results. In the study, a variant of the Advanced Message Queuing Protocol (AMQP) called RabbitMQ, also called as RMQ (Richardson, 2012; Videla & Williams, 2012) is utilized for real-time data transfer between the Raspberry Pi 3 and a post-processing device. Because each of the data collection, post-processing scripts, and visualization processing have to be performed continuously and sequentially, the RMQ has been used for data exchange between the processes to assist parallel data collection and processing. The processed data show an estimated distance of the object from the radar system in real-time, so that the system can support to monitor a certain area in a remote place if the two distant places are connected through a network.<div><br></div><div>This proposed radar system performed successfully to detect and track an object that was in the sight of the radar. Although further study to improve the system is required, the system will be highly suitable and applicable for research areas requiring sensors for exploration, monitoring, or surveillance because of its accessibility and flexibility. Users who will adopt this radar system for research purposes can develop their own applications that match their research environment for example to support robots for obstacle avoidance or localization and mapping.<br><div><div><div> </div> </div> </div></div></div></div></div>

No description available.

With the emergence of Convolutional Neural Network (CNN) models, precision of image classification tasks has been improved significantly over these years. Regional CNN (RCNN) model is proposed to solve object detection tasks with a combination of Region Proposal Network and CNN. This model improves the detection accuracy but suffer from slow inference speed because of its multi-stage structure. The Single Stage Detection (SSD) network is later proposed to further improve the object detection benchmark in terms of accuracy and speed. However, SSD model still suffers from high miss rate on small targets since datasets are usually dominated by medium and large sized objects, which don’t share the same features with small ones. On the other hand, geometric analysis on dataset images can provide additional information before model training. In this thesis, we propose several SSD-based models with adjusted parameters on feature extraction layers by using geometric analysis on KITTI and Caltech Pedestrian datasets. This analysis extends SSD’s capability on small objects detection. To further improve detection accuracy, we propose a two-stream network, which uses one stream to detect medium to large objects, and another stream specifically for small objects. This two-stream model achieves competitive performance comparing to other algorithms on KITTI and Caltech Pedestrian benchmark. Those results are shown and analysed in this thesis as well.