A vision system for UAV position control

In this paper, we propose a vision-based control system for the localization of an unmanned aerial vehicle (UAV) which equips on-board cameras. The proposed approach is combined a visual odometer with inertial measurement from IMU and pressure signal which compensate attitude and altitude, respectively. For the autonomous control of a UAV, we consider only natural landmarks provided by a feature tracking algorithm without the help of visual beacons or landmarks at known positions. The displacement of the UAV is estimated by the homography, the relationship between a pair of points. The experimental results show that it is possible to extract useful position information from the proposed vision system even when the UAV flies high, and the information can be used temporary position control of UAVs during GPS failure.