Design of a Multi-Tier Unmanned Aerial Vehicle Search Algorithm for Locating Radioactive Sources in a Large Complex Region

Year
2018
Author(s)
Benjamin Magocs - Institute for Nuclear Security at the University of Tennessee Knoxville
Dr. Matthew Cook - Institute for Nuclear Security at the University of Tennessee Knoxville
Josh Gerka - Institute for Nuclear Security at the University of Tennessee Knoxville
Abstract
By implementing a combination of techniques based on the search area, we are developing an algorithm that can rapidly search a large area to locate a radiological source. The algorithm has the ability to control an aerial vehicle to take initial measurements and develop an initial probability guess in addition to using a triangulation method to verify approximate locationsof specific sources. Once an initial estimate has been calculated, multiple UAS are controlled by the algorithm to use specified swarm patterns that are produced by optimizing the separation distances between the vehicles based on the useful detection radii of the radiological onboard detectors. Once further probability calculations have been made usingBayesian algorithms, the “hot spots” in the search region can be checked using a single direction rightangle turn method to pinpoint sources.