Robust collaborative 3D target localization using UAVs with state uncertainties

Abstract

This paper addresses the problem of target localization using measurements collected by a fleet of cooperative Unmanned Aerial Vehicles (UAVs). The positions and orientations of UAVs are partly unknown. The measurement noise and other uncertainties of the system are assumed to be unknown but bounded. Moreover, some of the UAVs of the feet may produce erroneous data (outliers). Several UAVs are assumed to observe a single target, to exchange the collected information. Then they have to characterize the set of all target locations consistent with the measurements, the bounded measurement noise, and the bounded uncertainty of the position and orientation of the UAVs while being robust to possible outliers. A set-membership approach based on set inversion via interval analysis is used. Thick intervals are required to account for the uncertainty in the location and orientation of the UAVs. The robustness to outliers is obtained using q-relaxed intersection.