|
|
|
||||||
|
|
|
HD4AR
Abstract:Construction site activities require access to large amounts of cyber-information, such as Building Information Models (BIM) and project specifications. Field personnel carry around large stacks of project documents, or frequently travel to a trailer to lookup this information. Recently, several context-aware techniques have been proposed to deliver relevant information to on-site users by intelligent interpretation of their environment. These techniques primarily rely on GPS and/or wireless localization, which typically does not provide sufficient precision in congested construction sites. To address these limitations, this paper extends our work on Hybrid 4-Dimensional Augmented Reality (HD4AR), a high-precision mobile augmented reality system that allows field personnel to query and access semantically-rich 3D cyber-information and see it overlaid on top of real-world imagery. With our proposed method, field personnel can use mobile devices to take pictures of building elements and be presented, on-site, with a detailed list of related cyber-information in an augmented reality (AR) format. In contrast to previous techniques, our proposed vision-based method localizes the user purely based on image matching and provides more accurate positioning. With HD4AR, the image captured by the field personnel using a mobile device is sent to a server to conduct GPU-based feature extraction and matching against pre-collected images from the jobsite. The mobile device’s 3D position and orientation is then accurately derived by solving for the Direct Linear Transform followed by a Levenberg-Marquardt optimization against an underlying Structure-from-Motion 3D point cloud model. The paper further validates the HD4AR localization method for several practical metrics. Particularly it presents the accuracy of GPU-based localization and further reduction in 3D reconstruction and localization time from our previous CPU-based method. The perceived benefits and limitations of the HD4AR system for on-site context-aware applications are discussed in detail. |
Bibliographic Info:
Hyojoon Bae, Mani Golparvar-Fard, & Jules White, ENHANCED HD4AR (HYBRID 4-DIMENSIONAL AUGMENTED REALITY) FOR UBIQUITOUS CONTEXT-AWARE AEC/FM APPLICATIONS, CONVR2012.Deployment Optimization for Avionics
Title: Deployment Optimization for Embedded Flight Avionics Systems
Abstract:Loosely-coupled publish/subscribe messaging systems facilitate optimized deployment of software applications to hardware processors. Intelligent algorithms can be used to refine system deployments to reduce system cost and resource requirements, such as memory and processor utilization. This article describes how we applied a computer assisted deployment optimization tool to reduce the required processors and network bandwidth consumption of a legacy flight avionics system. |
 |
Bibliographic Info:
Brian Dougherty, Jules White, Douglas C. Schmidt, Russell Kegley, Jonathan Preston, Deployment Optimization for Embedded Flight Avionics Systems, CrossTalk Journal, (to appear). This research has been funded in part by a grant from the Air Force Research Laboratories.
Smartphone Computing in the Classroom
Title: Smartphone Computing in the Classroom
Abstract:Smartphones, such as the iPhone and Google Android, have become extremely popular and constitute an ever-increasing share of computing platforms. For example, in the third quarter of 2010, 88.3 million PCs were sold worldwide. In that same quarter, 80 million smartphones were purchased. Roughly 20 million Android devices and 14.1 million iOS phones were sold. Moreover, whereas there was only a 7.6% growth in PC sales from the previous quarter, smartphones platform sales experienced approximately 30% growth. The processing capabilities of these smartphone devices have generated significant interest in—and development of—third-party applications. Most notably, the Apple application store contains more than 100,000 applications and has had more than 1 billion downloads to iPhones. It reached the 500 million download mark in January 2009 and the billion download mark by April 2009. These devices possess impressive hardware capabilities, as well as powerful software distribution, upgrade, and maintenance platforms supported by these application stores. Already, many of these applications possess cyber- physical1 qualities and supporting cloud services.2 For example, Google’s Latitude service uses a client- side application to capture location information from GPS sensors and intelligently aggregates information through a centralized cloud service that provides features, such as alerting you when you are near other friends that use Latitude. Given the increasing use of and excitement surrounding smartphone computing platforms, a key question is how they can be leveraged to enhance computing education. Here, we describe the ECE 4564 Network Application Design course at Virginia Tech, which we structured to use a combination of smartphone platforms, cloud computing, and open-ended assignments. We present the challenges that we faced when designing the course, our solutions to those challenges, and data measuring independent learn- ing outside of the classroom. Our experience teaching ECE 4564 and the results of our analyses of student learning show that smartphones have significant potential for improving self-directed student learning outside the classroom. |
 |
Bibliographic Info:
Jules White, Hamilton Turner,Smartphone Computing in the Classroom, IEEE Pervasive Computing, April-June, 2011
Smartphone Traffic Accident Detection
Title: WreckWatch: Automatic Traffic Accident Detection and Notification with Smartphones
Abstract:Traffic accidents are one of the leading causes of fatalities in the US. An important indicator of survival rates after an accident is the time between the accident and when emergency medical personnel are dispatched to the scene. Eliminating the time between when an accident occurs and when first responders are dispatched to the scene decreases mortality rates by 6%. One approach to eliminating the delay between accident occurrence and first responder dispatch is to use in-vehicle automatic accident detection and notification systems, which sense when traffic accidents occur and immediately notify emergency personnel. These in-vehicle systems, however, are not available in all cars and are expensive to retrofit for older vehicles. This paper describes how smartphones, such as the iPhone and Google An- droid platforms, can automatically detect traffic accidents using accelerometers and accoustic data, immediately notify a central emergency dispatch server after an accident, and provide situational awareness through photographs, GPS coor- dinates, VOIP communication channels, and accident data recording. This paper provides the following contributions to the study of detecting traffic accidents via smartphones: (1) we present a formal model for accident detection that com- bines sensors and context data, (2) we show how smartphone sensors, network connections, and web services can be used to provide situational awarenss to first responders, and (3) we provide empirical results demonstrating the efficacy of dif- ferent approaches employed by smartphone accident detection systems to prevent false positives. |
 |
Bibliographic Info:
Jules White, Chris Thompson, Hamilton Turner, Brian Dougherty, Douglas C. Schmidt, WreckWatch: Automatic Traffic Accident Detection and Notification with Smartphones, Springer Journal of Mobile Applications and Networks (to appear)