Wheeler, Jesse; Bean, Brennan; Schwartz, Sam; Christensen, Randy; Moon, Kevin (Utah State University)
Faculty Advisor: Moon, Kevin (College of Science, Mathematics and Statistics Department)
Unmanned aerial vehicles (UAV) often rely on GPS for navigation. GPS signals, however, are very low in power and are easily jammed or otherwise disrupted. Precise measurements of initial position and motion at the time of GPS signal loss would allow navigation for UAV navigation in GPS denied regions. This work presents a method for determining the navigation errors present at the beginning of a GPS-denied period by utilizing data from a synthetic aperture radar (SAR) system. This is accomplished by comparing an online-generated SAR image with a reference image obtained a priori. The distortions relative to the reference image are learned and exploited with a convolutional neural network to recover the initial navigational errors, which can be used to recover the true flight trajectory throughout the synthetic aperture. Our neural network approach outperforms traditional navigation recovery methods as well as other machine learning models.