Skip to content

About

A SLAM implementation combining FAST-LIO2 with pose graph optimization and loop closing based on Quatro and Nano-GICP

Resources

Stars

1 star

Watchers

0 watching

Forks

 
 

Latest commit

 

History

20 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FAST-LIO-SAM-QN


Computational complexity
in KITTI seq. 05 with i9-10900k CPU

  • FAST-LIO-SAM: max 118% CPU usage, 125 times of ICP, 124.9ms consumption on average
  • FAST-LIO-SAM-N (only Nano-GICP): max 164% CPU usage, 130 times of ICP, 61.9ms consumption on average
  • FAST-LIO-SAM-QN: max 247% CPU usage, 66 times of ICP, 942ms consumption on average
  • Note: loop_timer_func runs at fixed basic/loop_update_hz. So how many times of ICP occured can be different depending on the speed of matching methods.


KITTI seq 05 top view - (left): FAST-LIO2 (middle): FAST-LIO-SAM (bottom): FAST-LIO-SAM-QN


KITTI seq 05 side view - (top): FAST-LIO2 (middle): FAST-LIO-SAM (bottom): FAST-LIO-SAM-QN

Dependencies

  • C++ >= 17, OpenMP >= 4.5, CMake >= 3.10.0, Eigen >= 3.2, Boost >= 1.54
  • ROS
  • GTSAM
    wget -O gtsam.zip https://github.com/borglab/gtsam/archive/refs/tags/4.1.1.zip
    unzip gtsam.zip
    cd gtsam-4.1.1/
    mkdir build && cd build
    cmake -DGTSAM_BUILD_WITH_MARCH_NATIVE=OFF -DGTSAM_USE_SYSTEM_EIGEN=ON ..
    sudo make install -j16
  • Teaser++
    git clone https://github.com/MIT-SPARK/TEASER-plusplus.git
    cd TEASER-plusplus && mkdir build && cd build
    cmake .. -DENABLE_DIAGNOSTIC_PRINT=OFF
    sudo make install -j16
    sudo ldconfig

How to build and use

  • Get the code, build tbb first, and then build the main code. tbb is used for faster Quatro
    cd ~/your_workspace/src
    git clone https://github.com/engcang/FAST-LIO-SAM-QN --recursive
    
    # Note the directory
    cd FAST-LIO-SAM-QN/third_party/oneTBB && mkdir build installed
    cd build
    # Note the directory
    cmake .. -DCMAKE_INSTALL_PREFIX=../installed -DTBB_TEST=OFF -DCMAKE_BUILD_TYPE=Release
    make -j16 && make install
    
    cd ~/your_workspace
    # nano_gicp, quatro first
    catkin build nano_gicp -DCMAKE_BUILD_TYPE=Release
    # Note the option!
    catkin build quatro -DCMAKE_BUILD_TYPE=Release -DQUATRO_TBB=ON
    catkin build -DCMAKE_BUILD_TYPE=Release
    . devel/setup.bash
  • Then run (change config files in third_party/FAST_LIO)
    roslaunch fast_lio_sam_qn run.launch lidar:=ouster
    roslaunch fast_lio_sam_qn run.launch lidar:=velodyne
    roslaunch fast_lio_sam_qn run.launch lidar:=livox

Structure

  • odom_pcd_cb
    • pub realtime pose in corrected frame
    • keyframe detection -> if keyframe, add to pose graph + save to keyframe queue
    • pose graph optimization with iSAM2
  • loop_timer_func
    • process a saved keyframe
      • detect loop -> if loop, add to pose graph
  • vis_timer_func
    • visualize all (Note: global map is only visualized once uncheck/check the mapped_pcd in rviz to save comp.)

About

A SLAM implementation combining FAST-LIO2 with pose graph optimization and loop closing based on Quatro and Nano-GICP

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages