Vision-Based Runway Recognition and Position Estimation for UAV Autonomous Landing
B. Anbarasu, G. Fatima Shirly Anitha · SSRN Electronic Journal · 2016
Autonomous landing of a UAV on a runway is a challenging task. In this work runway recognition, position and attitude estimation of a UAV using vision based approach for autonomous landing of a UAV on a runway is presented. The UAV position and orientation relative to the runway is estimated using the geometry of the runway images. The width of the runway is a priori information required to estimate the altitude of the UAV above the runway. The runway can be identified in successive image frames using Scale Invariant Feature Transform (SIFT) based feature matching method. Once the runway is recognized the position estimation algorithm is triggered in the host computer and the position and lateral deviation of the UAV relative to the runway is estimated. The proposed algorithm is verified by interfacing with flightgear flight simulation software. Simulation results show the proposed algorithm is accurate and fast.