Vision based autonomous landing of an Unmanned Aerial Vehicle on a stationary target

The foremost, yet challenging problem of an Unmanned Aerial Vehicle (UAV) is the autonomous landing. This paper presents a solution for the independent landing of a quadrotor UAV on a stationary target. This method includes a searching and a landing part. The visual data is collected by the onboard camera of the AR Drone. Speeded Up Robust Features (SURF) method is used to detect and compute the keypoint descriptors and Fast Approximate Nearest Neighbor Search Library (FLANN) based descriptor matcher is utilized for the accurate matching of the template image with the captured images to determine the target position. The aim of this method is to minimize the distance between the drone and the target. A Proportional-Integral-Derivative (PID) controller is designed to generate the desired velocity signals for the drone. Also, in this work, the drone is directly navigating and landing on the target by adjusting the x, y and z velocities simultaneously. The method is tested and validated on a quadrotor with two different initial heights.