search

hovsg / HOV-SG

[RSS2024] Official implementation of "Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation"
https://hovsg.github.io
MIT License
186 stars 12 forks source link

RGB-D data collection in real world #4

Closed YESAndy closed 3 months ago

YESAndy commented 3 months ago

Hi,

Thank you for sharing this great work! I notice that you use Fast-LIO for localization which is very smooth in reconstructing the real-world building. I also try to use Fast-LIO from the official repo however the performance of the localization is not that good. Could you provide details about your modified version of Fast-LIO? Thanks!!

abwerby commented 3 months ago

Hello, We used FAST_LIO_LC to create the map of the real-world building, (get the RGBD sequence to create the graph). For the navigation tasks we used the real-world map PointCloud (FAST-LIO-LC) + Spot robot map (navigation stack) for the same building, to navigate autonomously. Our method can work with any SLAM system as long it accurate enough + Robot with a Navigation stack. You can take a look here: https://github.com/yanliang-wang/FAST_LIO_LC https://github.com/gisbi-kim/FAST_LIO_SLAM https://github.com/HViktorTsoi/FAST_LIO_LOCALIZATION