Computer Engineering Commons^™

Bayesian Optimization For Refining Object Proposals, Anthony D. Rhodes, Jordan Witte, Melanie Mitchell, Bruno Jedynak

Melanie Mitchell

We develop a general-purpose algorithm using a Bayesian optimization framework for the efficient refinement of object proposals. While recent research has achieved substantial progress for object localization and related objectives in computer vision, current state-of-the-art object localization procedures are nevertheless encumbered by inefficiency and inaccuracy. We present a novel, computationally efficient method for refining inaccurate bounding-box proposals for a target object using Bayesian optimization. Offline, image features from a convolutional neural network are used to train a model to predict an object proposal’s offset distance from a target object. Online, this model is used in a Bayesian active search ...

Fast On-Line Kernel Density Estimation For Active Object Localization, Anthony D. Rhodes, Max H. Quinn, Melanie Mitchell

Melanie Mitchell

A major goal of computer vision is to enable computers to interpret visual situations—abstract concepts (e.g., “a person walking a dog,” “a crowd waiting for a bus,” “a picnic”) whose image instantiations are linked more by their common spatial and semantic structure than by low-level visual similarity. In this paper, we propose a novel method for prior learning and active object localization for this kind of knowledge-driven search in static images. In our system, prior situation knowledge is captured by a set of flexible, kernel-based density estimations— a situation model—that represent the expected spatial structure of the ...

Seasonal Warranty Prediction Based On Recurrent Event Data, Qianqian Shan, Yili Hong, William Q. Meeker Jr.

William Q Meeker

Warranty return data from repairable systems, such as vehicles, usually result in recurrent event data. The non-homogeneous Poisson process (NHPP) model is used widely to describe such data. Seasonality in the repair frequencies and other variabilities, however, complicate the modeling of recurrent event data. Not much work has been done to address the seasonality, and this paper provides a general approach for the application of NHPP models with dynamic covariates to predict seasonal warranty returns. A hierarchical clustering method is used to stratify the population into groups that are more homogeneous than the than the overall population. The stratification facilitates ...

Addressing Multiple Bit/Symbol Errors In Dram Subsystem, Ravikiran Yeleswarapu, Arun K. Somani

Arun Somani

As DRAM technology continues to evolve towards smaller feature sizes and increased densities, faults in DRAM subsystem are becoming more severe. Current servers mostly use CHIPKILL based schemes to tolerate up-to one/two symbol errors per DRAM beat. Multi-symbol errors arising due to faults in multiple data buses and chips may not be detected by these schemes. In this paper, we introduce Single Symbol Correction Multiple Symbol Detection (SSCMSD) - a novel error handling scheme to correct single-symbol errors and detect multi-symbol errors. Our scheme makes use of a hash in combination with Error Correcting Code (ECC) to avoid silent data ...

Provisioning On-Line Games: A Traffic Analysis Of A Busy Counter-Strike Server, Francis Chang, Wu-Chang Feng, Wu-Chi Feng, Jonathan Walpole

Jonathan Walpole

A poster that illustrates the client/server model employed by an multiplayer online game, focusing on bandwidth usage.

The Use Of Agricultural Robots In Weed Management And Control, Brian L. Steward, Jingyao Gai, Lie Tang

Brian L. Steward

Weed management and control are essential for the production of high-yielding and high-quality crops, and advances in weed control technology have had a huge impact on agricultural productivity. Any effective weed control technology needs to be both robust and adaptable. Robust weed control technology will successfully control weeds in spite of variability in the field conditions. Adaptable weed control technology has the capacity to change its strategy in the context of evolving weed populations, genetics, and climatic conditions. This chapter focuses on key work in the development of robotic weeders, including weed perception systems and weed control mechanisms. Following an ...

Transportation Safety Performance Of Us Bus Transit Agencies And Population Density: A Cross-Sectional Analysis (2008-2014), Ilker Karaca, Peter T. Savolainen

Ilker Karaca

The paper examines the transportation safety performance of transit agencies providing public bus service in the US by using data from the National Transit Database (NTD)

Uses NTD data for a seven-year period from 2008 to 2014 • 3,853 observations for 651 public transportation agencies in 50 states

Seven types of bus transit fatalities and injuries (including passengers, operators, pedestrians, bicyclists)

Main explanatory variable: urban density obtained from the US Census figures

Other explanatory variables: total agency revenue miles, unlinked passenger trips, agency fleet size, and urban population

Developing A Workflow To Integrate Tree Inventory Data Into Urban Energy Models, Farzad Hashemi, Breanna L. Marmur, Ulrike Passe, Janette R. Thompson

Farzad Hashemi

Building energy simulation is of considerable interest and benefit for architects, engineers, and urban planners. Only recently has it become possible to develop integrated energy models for clusters of buildings in urban areas. Simulating energy consumption of the built environment on a relatively large scale (e.g., such as a neighborhood) will be necessary to obtain more reliable results, since building energy parameters are influenced by characteristics of the nearby environment. Therefore, the construction of a 3-D model of urban built areas with detail of the near-building environment should enhance simulation approaches and provide more accurate results. This paper describes ...

Top-Down Construction Cost Estimating Model Using An Artificial Neural Network, Douglas D. Gransberg, H. David Jeong, Ilker Karaca, Brendon Gardner

Ilker Karaca

This report contains the information and background on top-down cost estimating using artificial neural networks (ANN)_to enhance the accuracy of MDT early estimates of construction costs. Upon conducting an extensive review of MDT’s budgeting and cost estimating efforts, and following a survey of agency experts on the identification of the most salient project attributes with the dual-objectives of low effort and high accuracy, a rational method for top-down variable selection is proposed. Selected variables were further tested in their explanatory power of construction costs through the application of two cost estimating methodologies—multiple regression and artificial neural network ...

Integrating Mathematics And Educational Robotics: Simple Motion Planning, Ronald I. Greenberg, George K. Thiruvathukal, Sara T. Greenberg

George K. Thiruvathukal

This paper shows how students can be guided to integrate elementary mathematical analyses with motion planning for typical educational robots. Rather than using calculus as in comprehensive works on motion planning, we show students can achieve interesting results using just simple linear regression tools and trigonometric analyses. Experiments with one robotics platform show that use of these tools can lead to passable navigation through dead reckoning even if students have limited experience with use of sensors, programming, and mathematics.

Real-Time Insight : Developing A Internal Library Data Dashboard, Travis Clamon

Travis Clamon

Library systems are becoming increasingly open, providing API’s that can deliver data in real-time. ETSU Library staff wanted a way to collect metrics from various systems and present it in one centralized dashboard. This presentation will cover our experiences with the project, including successes, challenges, and future goals.

Authentication Via Openathens: Implementing A Single Sign-On Solution For Primo, Alma, And Ezproxy, Travis Clamon

Travis Clamon

OpenAthens is a hosted identity and access management service that provides a streamlined solution for implementing single sign-on authentication. This presentation will outline the steps East Tennessee State University took to configure OpenAthens authentication across the Alma, Primo, and EZproxy platforms. We will give a brief overview of the internal configurations related to LDAP integration, allocating electronic resources, and selectively assigning permissions. Finally, we will share our experiences with OpenAthens including support, vendor adoption, and end user benefits.

Eliminating Data Redundancy: Our Solution For Database Discovery Using Alma/Primo, Jacob Kindle, Travis Clamon

Travis Clamon

East Tennessee State University recently adopted Alma & Primo and was suprised by the lack of an A-Z database discovery module. Frustrated by having to maintain electronic resources separately on our library website and in Alma, we embarked on a goal to eliminate redundancy and use Alma/Primo exclusively. This presentation will cover our entire workflow in both Alma & Primo and the issues we encountered along the way. I'll first go over our process in Alma including MARC record creation, electronic collection setup, and the top level collection module. Next, I'll cover our workflow in Primo including normalization rules ...

A-Z Database Discovery Using Alma: Eliminate Redundancy And Simplify Your Workflow, Travis Clamon

Travis Clamon

Frustrated by having to maintain an A-Z databases list separately on our library website and in Alma/Primo, East Tennessee State University embarked on a goal to eliminate redundancy by using Alma as our primary source of metadata for eResources. This presentation will cover our entire workflow and the issues we encountered along the way. I'll first go over our process in Alma including MARC record creation, electronic collection setup, and the top level collection module. Next, I'll cover our workflow in Primo including normalization rules, scoping, and PNX display. The last section will cover the Alma API ...

Investigating The Use Of Bayesian Networks For Small Dataset Problems, Anastacia Maria Macallister

Anastacia MacAllister

Benefits associated with machine learning are extensive. Industry is increasingly beginning to recognize the wealth of information stored in the data they are collecting. To sort through and analyze all of this data specialized tools are required to come up with actionable strategies. Often this is done with supervised machine learning algorithms. While these algorithms can be extremely powerful data analysis tools, they require considerable understanding, expertise, and a significant amount of data to use. Selecting the appropriate data analysis method is important to coming up with valid strategies based on the collected data. In addition, a characteristic of machine ...

Should Robots Prosecute And Defend?, Stephen E. Henderson

Stephen E Henderson

The Digital Persona And Trust Bank: A Privacy Management Framework, Sree Nilakanta, Kevin P. Scheibe

Kevin P. Scheibe

Recently, the government of India embarked on an ambitious project of designing and deploying the Integrated National Agricultural Resources Information System (INARIS) data warehouse for the agricultural sector. The system’s purpose is to support macro level planning. This paper presents some of the challenges faced in designing the data warehouse, specifically dimensional and deployment challenges of the warehouse. We also present some early user evaluations of the warehouse. Governmental data warehouse implementations are rare, especially at the national level. Furthermore, the motivations are significantly different from private sectors. Designing the INARIS agricultural data warehouse posed unique and significant challenges ...

The Subject Librarian Newsletter, Engineering And Computer Science, Spring 2018, Buenaventura "Ven" Basco

Buenaventura "Ven" Basco

No abstract provided.

The Subject Librarian Newsletter, Engineering And Computer Science, Fall 2017, Buenaventura "Ven" Basco

Buenaventura "Ven" Basco

No abstract provided.

Learning To Love Data (Week): Creating Data Services Awareness On Campus, Katie Wissel, Lisa Deluca

Lisa DeLuca, MLIS, MPA

No abstract provided.

Impact Of The Application Layer Protocol On Energy Consumption, 4g Utilization And Performance How The Application Layer Protocol Strengthens Green Iot, Sachin Pawaskar, Johannes Hofer

Sachin Pawaskar

The 2018 Nvidia Ai City Challenge, Milind Naphade, Ming-Ching Chang, Anuj Sharma, David C. Anastasiu, Vamsi Jagarlamudi, Pranamesh Chakraborty, Tingting Huang, Shuo Wang, Ming-Yu Liu, Rama Chellappa, Jenq-Neng Hwang, Siwei Lyu

David C. Anastasiu

Semiactive Virtual Control Method For Robots With Regenerative Energy-Storing Joints, Hanz Richter, Daniel J. Simon, Antonie J. Van Den Bogert

Antonie J. van den Bogert

A framework for modeling and control is introduced for robotic manipulators with a number of energetically self-contained semiactive joints. The control approach consists of three steps. First, a virtual control design is conducted by any suitable means, assuming a fully-actuated system. Then, virtual control inputs are matched by a parameter modulation law. Finally, the storage dynamics are shaped using design parameters. Storage dynamics coincide with the system's internal dynamics under exact virtual control matching. An internal energy balance equation and associated self-powered operation condition are given for the semiactive joints. This condition is a structural characteristic of the system ...

Power-Stepped Protocol: Enhancing Spatial Utilization In A Clustered Mobile Ad Hoc Network, Chansu Yu, Kang G. Shin, Ben Lee

Chansu Yu

While most previous studies on mobile ad hoc networks (MANETs) rely on the assumption that nodes are randomly distributed in the network coverage area, this assumption is unlikely to hold, as nodes tend to be cluttered around hot spots like the site of an accident or disaster. We refer to this as a clustered layout. Intuitively, a MANET with the clustered layout may suffer from serious performance degradation due to the excessive collisions in congested hot spots and space underutilization of sparse areas. In this paper, we propose a power-controlled network protocol, called the power-stepped protocol (PSP), that maximizes the ...

Many-To-One Communication Protocol For Wireless Sensor Networks, Chansu Yu, Robert Fiske, Seungmin Park, Won-Tae Kim

Chansu Yu

This paper proposes a novel communication protocol, called Many-to-One Sensors-to-Sink (MOSS), tailored to wireless sensor networks (WSNs). It exploits the unique sensors-to-sink traffic pattern to realize low-overhead medium access and low- latency sensors-to-sink routing paths. In conventional schedule-based MAC protocols such as S-MAC, sensor nodes in the proximity of the event generate reports simultaneously, causing unreliable and unpredictable performance during a brief but critical period of time when an event of interest occurs. MOSS is based on time division multiple access (TDMA) that avoids energy waste due to collisions, idle listening and overhearing and avoids unreliable behavior mentioned above. A ...

Cache Invalidation Strategies For Internet-Based Vehicular Ad Hoc Networks, Sunho Lim, Chansu Yu, Chita R. Das

Chansu Yu

Internet-based vehicular ad hoc network (Ivanet) is an emerging technique that combines a wired Internet and a vehicular ad hoc network (Vanet) for developing an ubiquitous communication infrastructure and improving universal information and service accessibility. A key design optimization technique in Ivanets is to cache the frequently accessed data items in a local storage of vehicles. Since vehicles are not critically limited by the storage/memory space and power consumption, selecting proper data items for caching is not very critical. Rather, an important design issue is how to keep the cached copies valid when the original data items are updated ...

The Case For It Training Within Guinea’S Ministry Of Agriculture: Evaluating Performance And Usability, Simon Cleveland

Simon Cleveland, Ph.D.

Hardware-Based Restricted Boltzmann Machine Implementation Using Probabilistic Spin Logic Devices, Adedoyin Adepegba

Adedoyin Adepegba

No abstract provided.

Work In Progress: Teaching Game Design And Robotics Together: A Natural Marriage Of Computing And Engineering Design In A First-Year Engineering Course, Adam R. Carberry, Ashish Amresh

Ashish Amresh

The increased dependence on computer programming in engineering has made it essential for engineering students to learn about programming throughout their undergraduate education. In the same vein, computing students benefit when given an opportunity to learn more about engineering design and systematic thinking. This paper discusses how one college embedded computing and engineering into a combined first-year introductory course. The course fuses computing and engineering using game design and robotics as an offering for both cohorts of students to work together in a multidisciplinary environment. Over the course of the semester, students learn introductory computing and engineering design concepts by ...

Scrumtutor: A Web-Based Interactive Tutorial For Scrum Software Development, Sindhura Potineni, Srividya K. Bansal, Ashish Amresh

Ashish Amresh

In a traditional software engineering class, students are typically engaged in theoretical lectures followed by homework assignments or a project. Use of hands-on training and laboratory activities using realworld projects is more likely to produce students with a higher level of achievement and more confidence in the course material. If every topic or technique introduced in the course has a corresponding hands-on activity that demonstrates an application or use of the concept in the industry, students better understand the need for the technique and the learning environment is more interactive, engaging, and interesting to students. This paper presents a project ...