Geolocation of Unmanned Aerial Vehicles in GPS-Degraded Environments

This is the authors' abstract. We don't add key points for this paper.

Unmanned systems have become popular for applications such as reconnaissance, surveillance, and even as weapon systems, thus increasing the need to navigate these systems reliably through a variety of environments. For most of these type applications, there is a requirement that the unmanned system be able to determine its location on the earth. Typically, this is accomplished by using the Global Positioning System (GPS), and although GPS is certainly an integral part of modern navigation, there are some instances when it becomes degraded either by the environment or by jamming. Some of the environments that can cause degradation in GPS accuracy include under tree canopies, inside cities between skyscrapers (urban canyons), indoors, and in heavily mountainous terrains. The problem with each of these areas is that GPS signals from some satellites may become attenuated due to the previously mentioned environmental conditions, thus reducing the number of available satellites from which the position may be multilaterated. Also, multipath can produce inaccuracies when in environments with water, glass, or metal, because the GPS signals tend to be reflected causing errors in the distance calculations. In addition to degradation caused by these environmental effects, there are also methods of jamming GPS signals, which can also result in GPS being unavailable or of reduced quality. Because of these problems with GPS, current research is focusing on a method to potentially serve as a complement to the GPS system, particularly for unmanned ground vehicles (UGV) and unmanned aerial vehicles (UAV). This paper provides a description of a method that utilizes a mobile ad-hoc network of nodes, each of which are capable of obtaining ranges to surrounding nodes. The algorithms being studied assume the presence of such ranges, but are independent of the specific ranging technique being utilized, although it would be well suited for a Radio Frequency (RF)-based ranging method. These ranges, along with the positions of some nodes that know their position either by using GPS or another geolocation method, are then used by multilateration techniques to determine the location of the other nodes in the network. While similar methods have been previously implemented, this technique concentrates on ensuring that the network can truly be considered a mobile adhoc network in that all nodes may act as either ranging nodes or receivers, and no nodes are required to be in a fixed position.