Visualized in Foxglove: UZH-FPV Drone Racing Dataset

“Are we ready for autonomous drone racing? The UZH-FPV drone racing dataset”
In robotics, particularly in high-speed drones and humanoids, accurate and rapid state estimation and motion response are critical. However, aggressive trajectories pose challenges for state estimation due to large accelerations, rotations, and the apparent motion in vision sensors.
This visualization involves streaming the data after it has been imported into Foxglove, where IMU data streams at incredibly fast rates, ranging from 10 to 1000 Hz. This high-frequency data streaming is invaluable in robotics development, as it allows for precise correlation between high-resolution motion and perception data, enabling effective state estimation and responsive motion control.
It also displays other sensor information from a Qualcomm Snapdragon Flight board, ground truth from a Leica Nova MS60 laser tracker, as well as event data from an mDAVIS 346 event camera, and high-resolution RGB images from the pilot’s first-person view (FPV) camera.
This dataset is provided by the research of Jeffrey Delmerico, Titus Cieslewski, Henri Rebecq, Matthias Fässler, and Davide Scaramuzza
Link to the dataset and code:
fpv.ifi.uzh.ch/
github.com/uzh...