Autonomous Visual Navigation for a Flower Pollination Drone

In this paper, we present the development of a visual navigation capability for small drone enabling it to autonomously approach flowers. This is very important step towards fully autonomous flower pollinating nanodrone. The developed totally and relies its on on-board color camera, complemented with one simple ToF distance sensor, detect flower. proposed solution uses DJI Tello carrying Maix Bit processing board capable running all deep-learning-based image algorithms on-board. We two-stage servoing algorithm that first highly optimized object detection CNN localize flowers fly it. second phase, approaching flower, implemented by direct steering CNN. enables any in neighborhood, steer make drone’s rod touch trained deep learning models based an artificial dataset mix images real flowers, (synthetic) virtually rendered Our experiments demonstrate technically feasible. able detect, autonomously. 10 cm sized prototype sunflowers, but methodology presented paper can be retrained type.