Pose estimation of autonomous dirigibles using artificial landmarks

Abstract

Autonomous dirigibles - aerial robots that are blimps controlled by computer based on information gathered by sensors-are a new and promising research field in robotics, offering several possibilities for original research. One of them is the study of visual navigation for this kind of vehicle, which would fill an operational gap left by inertial, GPS and other kinds of well-established dead-reckoning navigation. In this paper, a simple pose estimation system based on artificial landmarks is introduced for the accomplishment of that task. The vision system is able to track artificial visual beacons-objects with known geometrical properties-and, from them, a geometrical methodology can extract information about the orientation and position of the blimp. A simple control system uses that data to keep the dirigible on a programmed trajectory. This visual control can be done using both fixed, external cameras focusing on the blimp, as well as on-board cameras, thus opening a way for future research on visual cooperation between aerial and ground mobile robots. Preliminary experimental results showing the correct functioning of the system are presented and discussed.