GAIA lab is focused on producing research that will help contribute to geospatial knowledge. Research projects range from analysis of the campus air quality environment to modelling spatial risk of roads. One of the main research goals for all GAIA projects is collaboration, which allows us to utilize spatial concepts and technology in an assortment of other fields. This process fosters communication and acheivement for students, and broadens their horizons through interdisciplinary study.
GAIA Lab has three main project umbrellas that the majority of research falls into. The founding research and flagship project of GAIA lab is SmartCampus, which is highlighted below. GAIA also sponsors various side projects that master students seek to accomplish.
Various projects are detailed further in their descriptions. See their profiles below.
Smart Campus Umbrella
The flagship project of GAIA lab is a collaboration among GIS, Geosciences, Facility Management, and the Office of Information Technology to build a geospatial cyberinfrastructure with GIS databases, geospatial cloud services, sensor networks, and web and mobile applications for a cyber-physical-social connected smart UTD campus.
he project is sub-divided into multiple facets to give students the ability to work more closely to research interests.
Smart Campus GeoDatabase
The foundation of smart campus
converting CAD As-Builts to Smart GIS Data
3D Interior GIS
Rendering campus interiors in 3D
for visualization and 3D analysis
Structure from Motion – Campus Exteriors
Creating realistic views of campus buildings for 3D rendering
Wayfinding and triangulation in the interior of campus buildings
Mobile Applications
A key future work
that will bridge SmartCampus and the campus community.
AQ Monitor – UTD Campus
A subset of SmartCampus and AQTEXAS
Real time air quality monitoring on campus
Remotely Sensed Imagery Repository
Using UAS (drones) we plan on building a dataset of remotely sensed imagery for campus use.
Reverse Viewshed for UAS LOS
In part of FAA guidelines, in order to fly a UAS on campus this project determines visual observer locations
SAFE-NET Umbrella
“SAFE-NET: An Integrated Connected Vehicle and Computing Platform for Public Safety Applications is a partnership between Southern Methodist University, University of Texas at Dallas, Ericsson North America (Ericsson), Southwest Research Institute (SwRI), and Dallas Fire-Rescue Department. The project seeks to develop an integrated communication and computational platform to support efficient and safe equipment and personnel mobilization for emergency response.” – NIST
A larger description of the project can be found on https://www.nist.gov/ctl/pscr/safe-net-integrated-connected-vehicle-computing-platform
The UT-Dallas contribution of the research so far has estimated the risk of 100m street segments, as well as begun to explore emergency response route patterns and the variables that affect them.
Click learn more below to see these projects in depth.
Road Network Risk Modelling
Logistic regression and random forest classification are used to predict accident likelihood on 100m segments at 1-hour intervals in Dallas, TX, using explanatory variables including time, the space syntax variables of integration and choice, and other site and dynamic characteristics.
AVL – Emergency Response Vehicle Analysis
An analysis of emergency vehicle trajectories to examine relationships among dispacthed units, routes taken to emergency situations, accidents occuring en route, and whether the vehicle made it to the destination. Preliminary data cleaning demonstrates the challenges of temporal GIS data. Further updates will be provided as the project continues.
Other Projects
GAIA lab conducts other research that does not fall into the previous two umbrellas. Below are some unique projects that the team has conducted.
Determining What Impacts Diet
A Master’s project by Dylan Campbell, which examines factors that influence people’s diet. The study utilizes the powers of GIS to analyze spatial relationships in a qualitative fashion.
Geographic Event Modelling, Human Dynamics of Baltimore, MD
This study examined the spatio-temporal distribution of social events in Baltimore, MD. We seeked to understand space & time relationships among event pairs by utilizing co-occurence and co-location algorithms. The tools developed in this study can be applied to other study areas. The data came from Eventful.com through API access and demonstrate examples of Big Data.