We disclosed AERO, the design of our surveillance drone with embedded AI for real-time object detection and tracking, in our recently accepted Remote Sensing journal paper.

AERO is a #UAV brain system with onboard #AI capability using #GPU-enabled edge devices. It encompasses a novel multi-stage deep learning module that combines #objectdetection (#YOLOv4 and #YOLOv7) and tracking (#DeepSort) with #TensorRT accelerators to capture objects of interest with high accuracy and transmit data to the #cloud in real-time without redundancy. AERO processes the detected objects over multiple consecutive frames to maximize detection accuracy. The experiments show a reduced false positive rate (0.7%), a low percentage of tracking identity switches (1.6%), and an average inference speed of 15.5 FPS on a Jetson Xavier AGX edge device.

The paper is available online: here