This is the implementation of our paper: "Building Maps for Autonomous Navigation Using Sparse Visual SLAM Features" in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017. Our implementation is built on the top of the open-source stereo SLAM ORB-SLAM2. It runs in ROS (developed in version of indigo). Eigen, OpenCV, and CGAL are needed.
-
CGAL lib (mapping中的 triangulation用到) sudo apt-get install libcgal-dev sudo apt-get install libcgal-qt5-dev
-
修改g2o eigen库的错误 typedef Eigen::PermutationMatrix<Eigen::Dynamic, Eigen::Dynamic, SparseMatrix::StorageIndex> PermutationMatrix;
The high resolution video of our submitted paper is: https://1drv.ms/v/s!ApzRxvwAxXqQmRtBPbu_Q27V5o4f (720p is recommmanded)
To run this code, complie it with ROS in linux system, and type:
rosrun light_mapping ros_stereo_kitti
It will read the configuration file in the config folder, such as KITTI05.yaml. Please adjust it accordingly to your environment.
The source code is released under GPLv3 license. If you use our code, please cite our paper:
[1] Yonggen Ling and Shaojie Shen, "Building Maps for Autonomous Navigation Using Sparse Visual SLAM Features" in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2017. For more questions, please contact ylingaa at connect dot ust dot hk .