Towards Vision-Based Autonomous Landing for Small UAVs - First Experimental Results of the Vision System

Abstract

This paper describes the current development of intelligent vision capabilities for small sized Unmanned Aerial Vehicles (UAV) as part of the program “CAROLO” at the Technical University of Braunschweig, Germany. The objective of this development is to create image processing algorithms and appropriate hardware that suits the requirements of small sized UAVs. Enabling vision-based flight guidance during landing was chosen as the first task and acts as a technology demonstrator for future vision capabilities. The characteristics of a laid out marker are extracted from the on-board imagery. Relative position and attitude of this marker relative to the body-fixed coordinate system are determined. This concept conveys information about the designated landing point to the aircraft. An experimental installation of the vision system was used for first measurements. This experiment and its results – in particular the attainable accuracies of the presented vision system, are the subject of this publication.