Towards a ground navigation system based in visual feedback.pdfVIP

Proceedings of the 2012 IEEE Intelligent Vehicles Symposium Workshops Towards a ground navigation system based in visual feedback provided by a mini UAV ´ ˜ Mario Garzon , Joao Valente, David Zapata, Rigoberto Chil and Antonio Barrientos Abstract— This paper addresses initial efforts to develop a robot navigation is supported by visual data from the aerial navigation system for ground vehicles supported by visual robot. feedback from a mini aerial vehicle. A visual-based algorithm One of the advantages of using an aerial visual navigation computes the ground vehicle pose in the world frame, as well as possible obstacles within the ground vehicle pathway. Relying system is that the UGV field of view (FOV) is dynamic. on that information, a navigation and obstacle avoidance system This means that by controlling the UAV height and relative is used to re-plan the ground vehicle trajectory, ensuring position, the UGV at some stage can do a pseudo-zoom on an an optimal detour. Finally, some experiments are presented obstacle or any interesting object, as well as reconnaissance employing a unmanned ground vehicle (UGV) and a low cost of areas outside the UGVs FOV. Moreover, it is also possible mini unmanned aerial vehicle (UAV). to identify rugged terrain, floor openings or negative ob- I. INTRODUCTION stacles, and other unexpected navigation obstructions in the UGV surrounds. This collaboration ensures the UGVs safety Mobile robotic syste