Date of Award


Document Type


Degree Name

Master of Science in Computer Science


Department of Electrical and Computer Engineering

First Advisor

John F. Raquet, PhD


This research involves the design, implementation, and testing of a high-speed, real-time kinematic, precise differential GPS positioning system for use in airborne applications such as automated aerial-refueling and close formation flying. Although many of the current ambiguity resolution techniques use the residuals from the least squares position estimation to determine the true ambiguity set, this thesis presents a novel approach to the ambiguity resolution problem, called the minimum indicator. Instead of assuming the ambiguity set with the lowest residuals is the true set, other special characteristics of the residuals are examined. This increases the confidence that the algorithm has selected the true ambiguity set. The end result was the first-ever successful in-flight demonstration of close formation flight, culminating in over 11 hours of close formation flying with a mean radial spherical error of 3.3 centimeters (0.108 feet). Other areas addressed include: the difference between "pre-fit" and "post-fit" residuals in the conditional probability calculation, the impact of a simplified dynamics model on system performance, the effect of widelane observables on the time-to-fix the correct double-difference ambiguities, and dynamically adjusting the time constant and standard deviation of relative acceleration states in the FOGMA model.

AFIT Designator


DTIC Accession Number