Autonomous Flight in GPS-Challenging Environments Exploiting Multi-UAV Cooperation and Vision-aided Navigation

This paper presents an algorithm that allows an Unmanned Aerial Vehicle (UAV) to safely fly in GPS-Challenging environments exploiting cooperative and vision-based navigation techniques. In particular, a sensor fusion approach based on the Extended Kalman Filter (EKF) is presented, in which measurements from on-board inertial sensors and magnetometers are integrated with information provided by cooperative UAVs (namely GPS-based position), vision-based tracking, and monocular pose estimation. The performance of the proposed approach has been evaluated on the basis of numerical simulations. Achieved results show the potential of combining both multi-UAV and visionbased navigation approaches where one technique can compensate for the limits of the other, improving the overall navigation performance.