PLOS One just published Autonomous drone hunter operating by deep learning and all-onboard computations in GPS-denied environments
https://t.co/jClizDctJF
#dronehunter#dronesforgood#uav#Robots
On the @NVIDIAEmbedded Jetson TX2 with a 256×256 px input image, Tiny YOLO (@pjreddie) attained 76% accuracy in mAP measured at 0.5 IOU with an average frame rate of 8.55 fps: a 63% speed increase over the RGB-version of the same model.
#PLOSONE#YOLO https://t.co/LDhIT4zIBV
@pjreddie We used Darknet Tiny YOLO in our project to track drones. Check out our most recent publication. #PLOSONE: Autonomous drone hunter operating by deep learning and all-onboard computations in GPS-denied envir ... https://t.co/LDhIT4zIBV
Need a drone hunter?! @Gatwick_Airport & @HeathrowAirport maybe a small investment in furthering academic #uav research could solve shutdown problems in the future? Check out this recent publication @PLOS https://t.co/jClizDctJF
Our autonomous drone hunter uses an @AuvideaGmbH J120 carrier board to host a @NVIDIAEmbedded Jetson TX2. This allows us to directly connect the @Stereolabs3D#ZEDCamera to the Jetson. @PLOS published https://t.co/kpXPPyKKLP two days ago
This publication @PLOS https://t.co/kpXPPyKKLP uses a @NVIDIAEmbedded Jetson TX2 to power an autonomous hunting drone that hunts another drone in a GPS-deprived environment. The drone performs all computations onboard. #Drones#uav#ColumbiaUniversity