Optimal formation of unmanned aerial vehicles for minimizing localization error — A two-dimensional case

Abstract

Unmanned aerial vehicle (UAV) is attractive for localization where global positioning system (GPS) satellites are uncapturable and wireless anchor nodes are unavailable. The advantage of localization using UAVs is their movability, that is, even when only few UAVs are available, if each of them ranges a target node many times from different locations, the localization accuracy can be improved, because it virtually increases the number of UAVs, namely, anchor nodes. Unlike conventional localization systems, UAVs are movable, so they can be made a formation, such as a linear formation or a circular formation, to achieve higher localization accuracy for an individual target node to be localized. However, unfortunately, how to make the optimal formation of UAVs has been neither discussed nor solved. In this paper, we consider the formation problem of UAVs for localization. Assuming a two-dimensional time of arrival (TOA)-based localization, we theoretically derive the optimal formation of UAVs which can minimize the localization error.