VDB-EDT is an efficient and robust library for large-scale occupancy grid mapping and Euclidean distance transform. Compared with the state-of-the-art method, VDB-EDT can provide high-quality and more complete mapping results, while consuming less memory and processing time.
This package is developed based on ROS Kinetic and the Eigen3 library is also needed.
Dependencies for OpenVDB and the Catkin wrapper
sudo apt-get install libglfw3-dev libblosc-dev libopenexr-dev
sudo apt-get install liblog4cplus-dev
Download this package and go to the VDB-EDT folder
cd src
catkin_init_workspace
cd ..
catkin build
(currently not support catkin_make
command )
Please click here to download the cow-and-lady dataset.
Please click here to download the large-scale SubT dataset (background grid is with a size of 50m*50m).
Make sure the current path is at VDB-EDT folder, i.e., in the catkin workspace
soource devel/setup.bash
roslaunch vdb_edt VDB-EDT.launch
Play the cow-and-lady dataset. There exist a severe synchronization problem in this dataset. To sufficiently process this dataset, a play rate of 0.1 is recommended
rosbag play data.bag -r 0.1
After finishing testing cow-and-lady dataset, play the SubT dataset
rosbag play SubT.bag
Results on cow-and-lady dataset
The VDB Ros Wrapper inside this package is developed by Rohit Garg from Air Lab CMU based on the work of ETHZ ASL. We give our special acknowledgement to him.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。