Date of Award
Master of Science in Computer Engineering
Department of Electrical and Computer Engineering
Michael J. Mendenhall, PhD
Much of the recent successes in the Iraqi theater have been achieved with the aid of technology so advanced that celebrated journalist Bob Woodward recently compared it to the Manhattan Project of WWII. Intelligence, Surveillance, and Reconnaissance (ISR) platforms have emerged as the rising star of Air Force operational capabilities as they are enablers in the quest to track and disrupt terrorist and insurgent forces. This thesis argues that ISR systems have been severely under-exploited. The proposals herein seek to improve the machine-human interface of current ISR systems such that a predictive battle-space awareness may be achieved, leading to shorter kill-chains and better utilization of high demand assets. This thesis shows that, if a vehicle is being tracked by an ISR platform, it is possible to predict where it might go within a Time Horizon. This predictive knowledge is represented graphically to enable quick decisioning. This is accomplished by using Geo-Spatial Information Systems (GIS) obtained from municipal, commercial, or other ISR sources (e.g., hyperspectral) to model an urban grid. It then employs graph-theoretic search algorithms that prune the future state-space of that vehicle's environment, resulting in an envelope that constricts around all possible destinations. This thesis demonstrates an 81 % success rate for predictions carried out during experimentation. It further demonstrates a 97 % improvement over predictions made solely with models based on vehicular motion. This thesis reveals that the predictive envelopes show immense promise in improving ISR asset management, offering more intelligent interdiction of targets, and enabling ground sensor-cueing. Moreover, these predictive capabilities allow an operator to assign assets to make precise perturbations on the battle-space for true event-shaping. Finally, this thesis shows that the proposed methodologies are easily and cost-effectively deployed over existing Air Force architectures using the Software as a Service business model.
DTIC Accession Number
Muster, Richard T., "Exploitation of Geographic Information Systems for Vehicular Destination Prediction" (2009). Theses and Dissertations. 2470.