My Master's thesis is entitled “Detection Likelihood Maps for Wilderness Search and Rescue: Assisting Search by Utilizing Ground Searcher GPS Track Logs.”
The goal of this research is to provide ground searchers and search managers with a visual representation of the quality of search coverage.
Completed tasks:
Implemented CPU and CUDA viewshed library
Created simulation environment in the C4 game engine
Implemented the ability to collect Distance to Visual Obstruction (DTVO) and Average Maximum Detection Range (AMDR) values
Current tasks include:
Create touch screen application that displays the contour map, detection likelihood maps, and lets the user place markers on identified objects (this is because the Google Earth plugin keeps crashing)
Fully populate the simulation environment with desired vegetation
Collect detection data used to build the detection likelihood Bayesian model
Create a Bayesian model for learning the detection likelihoods
Design a secondary auditory task for the user study
Validate accuracy of detection likelihood model in field experiments
Switching to Dr. Seamon's Internet Security lab upon completion of Masters.
Cutting-edge technology
Computer and internet security
Parallel processing, especially using CUDA
Brain-computer interfaces
The great outdoors: backpacking, fishing, etc.
L. Lin, M. Roscheck, M. A. Goodrich, and B. S. Morse. Supporting Wilderness Search and Rescue with Integrated Intelligence: Autonomy and Information at the Right Time and the Right Place. In Twenty-Fourth AAAI Conference on Artificial Intelligence, Special Track on Integrated Intelligence. July, 2010, Atlanta, Georgia, USA.