Key facts
Contents & fields
The dataset contains over 30 km of visual-inertial flight data collected by a DJI M600 UAV with a gimbaled camera. Each location includes Teach and Repeat runs (reverse direction). Data includes Google Earth images, human-readable text files, and a set of Python tools. No field-level description is provided on the source page.
Research uses
Suitable for research in UAV visual localization including robustness to perspective change, lighting change and seasonal changes, as well as satellite image UAV navigation, experience-based localization and simultaneous localization and mapping.
This card was drafted from the source page; institution, coverage, time span, scale, fields and license are subject to the official page (pending human review).
Keywords
Why this is hard to get on your own
Research on UAV visual localization often requires long-term outdoor data across diverse environments and conditions, but public datasets are often limited in scope or scale.