Vision-based Cooperative Moving Path Following for Fixed-Wing UAVs

This paper addresses the problem of collaborative ground target tracking by a group of fixed-wing Unmanned Aerial Vehicles (UAVs) using vision in the loop. The UAVs adopt a circular path formation centered at the target’s coordinates and move together with it, using the Moving Path Following (MPF) method. A distributed control architecture is implemented, where each vehicle obtains the telemetry data from the preceding vehicle, through a chained communication network. A computer vision system based on machine learning techniques is proposed to close the cooperative MPF control loop. The obtained results show the efficiency of the proposed control system in realistic software-in-the-loop simulations.