Vision-based method for autonomous landing

Abstract
This paper presents eight visual cues to be used as part of a control loop for autonomous landing. The idea is based on fixating the camera at the vanishing point of the projection of the runway in the image. Two-dimensional geometrical cues are used to derive the visual cues. The visual cues that are extracted are the relative location of the camera to the runway in terms of the runway width, orientation of the camera, and relevant angles such as glide slope angle.