作者: Jesus Pestana , Jose Luis Sanchez-Lopez , Pascual Campoy , Srikanth Saripalli
DOI: 10.1109/SSRR.2013.6719359
关键词: Vision based 、 Global Positioning System 、 Robustness (computer science) 、 Video tracking 、 Computer science 、 Drone 、 Image plane 、 Artificial intelligence 、 Image sensor 、 Computer vision 、 Visual servoing
摘要: We present a vision based control strategy for tracking and following objects using an Unmanned Aerial Vehicle. have developed image visual servoing method that uses only forward looking camera from multi-rotor UAV, without any dependence on GPS systems. Our proposed tracks user specified object continuously while maintaining fixed distance the also simultaneously keeping it in center of plane. The algorithm is validated Parrot AR Drone 2.0 outdoor conditions people, occlusions fast moving objects; showing robustness systems against perturbations illumination changes. experiments show system able to track great variety suburban areas, among others: windows, AC machines, cars plants.