*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Each year, GVU provides seed grants, with support from IPaT, to research initiatives committed to building on our success in interdisciplinary research and innovation in the human experience of computing. These investments create a path for external funding as our research prospers.
2012 Research and Engagement Grants
Computational Play
We currently have little understanding of how people and intelligent agents might engage in unstructured, co-creative domains, where each team member takes an equal part in applying creative problem solving within a social context. In this project, we will investigate the problem of co-creative intelligence in the domain of imaginary play. Imaginary play is a fundamental aspect of human existence. Specifically, we will target play between humans and robotic systems, focusing on those aspects of human-robot interaction that we believe are missing in current robotics research: robot as peer to the human instead of subservient, interactivity, open-ended tasks, and evolving social roles.
Team: Brian Magerko (LMC), Andrea Thomaz (IC), and Mark Riedl (IC)
DMITRI Data Analysis
The DMITRI project has built a unique, comprehensive data set capturing daily life and diabetes management information from adult diabetics. This data includes personal logs, nutritional logs, clinical history, questionnaire data, and data from a range of on-body monitoring equipment including insulin pump dosage logs, Dexcom continuous clucose minitor, SenseWear activity monitor with accelerometer, GSR, and skin temperature sensing, Polar heart monitor, Philips Actiwatch, and Zeo sleep monitor. In this project, we aim to address the "capture and access" challenges of analyzing this data by applying our expertise in pattern recognition and wearable sensing to identify correlations and patterns in the dataset, shedding light on the impact of personal behavior on diabetes management.
Team: Thad Starner (IC) and Nate Heintzman (Biomedical Informatics, UCSD Dept. of Medicine)
Georgia Tech Campus Driving Simulation and Research
Applying driving simulators for in-vehicle research allows for a wide range of studies to be performed particularly when investigating cognitive demand and distraction caused by devices in the car. By using simulations, researchers can investigate driving behaviors in high-risk situations without putting participants or others in harm’s way. Currently being conducted within the School of Psychology at Georgia Tech, in-vehicle research could provide more insight into behavior and increase in applicability if participants were able to drive in areas that they are familiar with. Specifically, research being done in coordination with the Atlanta Shepherd Center investigating the use of in-vehicle technologies to assist individuals who have had a Traumatic Brain Injury could benefit largely through these real location maps. The Georgia Tech School of Architecture coincidentally has already developed a 3D model of the Georgia Tech campus and some of the surrounding areas including the Peachtree corridor (26 miles along Peachtree Street). However, in order to make this model usable within the simulator, it must be optimized and converted in a compatible format. Researchers in the School of Architecture and School of Psychology will be working on creating methods and conversion processes that will allow any 3D model to be integrated into the simulator. Development of this process of conversion will allow Georgia Tech to offer documentation and map-creation services to other researchers around the world assisting in increasing the applicability of in-vehicle research.
Team: Bruce Walker (Psych/IC) and Racel Williams (Architecture)
Cycle Atlanta Crowd-sourced Bike Route Desirability
Fifty percent of all trips are 3 miles or less, yet only 1.8% of those trips are biked. Meanwhile, 35.7% of US adults are obese and the transportation sector accounts for 32% of US greenhouse gases. One of the main reasons citizens do not use the healthier mode of cycling is due to a lack of safe infrastructure—dedicated bicycle routes, roads with bicycle lanes, and other designated bicycle facilities. The City of Atlanta has a desire to put proper cycling infrastructure in place but needs better information from citizens about where they currently and would like to cycle. Therefore, the initial goal of the Crowd-sourced Bicycle Route Desirability project is to modify the open-source CycleTracks application (previously adopted in San Francisco, CA, and Austin, TX.) for use in Atlanta. CycleTracks tracks the existing routes of cyclists using their smart phones and allows comparison of these routes to the quickest path from origin to destination. This allows us to begin to make appropriate infrastructure improvements to the most traveled routes in a study area by seeing logical paths that cyclists avoid. A second phase of the project would develop applications allowing riders to express their desired bike routes even if they currently do not cycle because of lack of adequate facilities.
Team: Kari Watkins (Civil Engineering) and Chris Le Dantec (LMC)
