Flight Testing of Cell Network-Based Navigation for sUAS

Abstract

The operational weaknesses and limitations of a navigation system based solely on GPS have long posed a problem for unmanned aircraft. The lack of any non-GPS-based backup system leaves vehicles vulnerable to sources of signal degradation or loss such as GPS jamming and signal multipathing, which can pose a serious safety hazard. Previous work has been done to develop an alternate navigation system, which generates a position estimate based on signals from cellular and Wi-Fi networks, known as WiCeNav. This study initially attempted to integrate WiCeNav onto an Unmanned Aerial Vehicle, or UAV. However, this proved unfeasible due to the fact difficulty of obtaining sufficient Wi-Fi data in the test environment. An alternate position estimation solution using multilateration of cell signals was therefore devised, and numerous problems associated with prior attempts at developing a system of this type were revisited and investigated in detail, shedding light on many key aspects of the functioning of cellular networks. Flight testing was conducted to investigate the feasibility of using the system for autonomous navigation. This did not produce a positive result, in part due to inaccuracy associated with the model used to relate power received to distance form a given tower, and in part due to inherent limitations of the network itself. However, this research provided valuable insight into the challenges of using cell networks for navigation, and a number of modifications and improvements are suggested which may drastically improve the accuracy and robustness of the system.