UAV´s Automatic Landing in All Weather Based on the Cooperative Object and Computer Vision

The Global Positioning System (GPS) is interfered easily, for the need of Unmanned Aerial Vehicle (UAV)´s landing automatically and accurately in all weather, a new program of UAV´s automatic landing is proposed based on the cooperative infrared object on the runway and the infrared computer vision in UAV. The cooperative object is recognized by affine moment invariants that is robust to geometric deformation from different big bevel angle of view in distant place. As there are some distortions of camera lens, then we used the characteristic points from the image of the cooperative infrared object to compute UAV´s the angle of pitching, the angle of roll and the angle of yaw relative to the runway by Tsai calibration method that can rectify camera lens distortions. The experimental result shows that the average correct identification rate reaches 97%, better than the two prevenient algorithms based on the direction code and the traditional moment invariants, the time consumption of this new recognition algorithm is only 25% of the two algorithms. Simulations show that the method of pose estimation in this paper is full-informationed, accurate and feasible for the pose estimation of the UAV comparing with the results of references.