The overall system framework for large-scale 3D map building in partially GNSS-denied scenes, which consists of two operating modes: LiDAR-only mode and GNSS-LiDAR mode. While working in GNSS-denied scenes, LiDAR odometry runs with high frequency and outputs estimations with local registration errors, and LiDAR mapping provides more accurate pose estimations with low frequency.
While moving from GNSS-denied scenes to open scenes, 3D-M-Box is working in GNSS-LiDAR mode. The auto coordinate alignment algorithm is applied to align the coordinate system between LiDAR and GNSS by registering a group of poses obtained from LiDAR mapping and GNSS within a certain time window. Finally, GNSS-constrained LiDAR mapping outputs the pose and point clouds with high accuracy.

The picture shows that 3D-M-Box can accomplish online pose estimation and map building in GNSS-denied scenes.

ROS (tested with kinetic)
gtsam (Georgia Tech Smoothing and Mapping library, 4.0.0-alpha2)

wget -O ~/Downloads/gtsam.zip https://github.com/borglab/gtsam/archive/4.0.0-alpha2.zip   
cd ~/Downloads/ && unzip gtsam.zip -d ~/Downloads/    
mkdir build && cd build    
cmake ..   
sudo make install    

cd ~/catkin_ws/src  
git clone https://github.com/ZhuangYanDLUT/lidar_gnss_mapping.git  
cd ~/catkin_ws  
catkin_make  

roslaunch lidar_gnss_mapping lidar_gnss_mapping.launch
In second terminal play sample data from test.bag (the access code is ri4b)
rosbag play -s 35 test.bag