Multi-volume .7z archives (like .002 ) are used to manage the massive size of the dataset, which totals approximately 3 TB .

Research Draft: Analysis of the Tartan_Space Dataset Component

The file is likely a secondary part of a multi-volume compressed archive belonging to the TartanAir dataset . This dataset, developed by CMU's AirLab, is used to push the limits of Visual SLAM (Simultaneous Localization and Mapping) in challenging simulated environments.

The "Tartan_Space" sub-environment specifically focuses on sci-fi or extraterrestrial scenes .

TartanAir is a large-scale dataset for robot navigation collected in photo-realistic simulation environments.