A LiDAR-Inertial localization package based on FAST-LIO2, operating with a known prior map.

Fig 1 :Indoor test without manual input of the initial pose

Fig 2 :Initial pose given by ScanContext(left) and Localization detail(right)

FAST-LOCALIZATION is a framework for relocalization in a known map, based on FAST-LIO2. It employs ScanContext for initial global localization. Compared to FAST_LIO_LOCALIZATION, FAST-LOCALIZATION does not require manual input of the initial pose. By utilizing continuous global point cloud constraints, it achieves more stable and globally consistent accurate pose estimates.

Ubuntu >= 16.04

ROS >= Melodic. ROS Installation

PCL >= 1.8, Follow PCL Installation.

Eigen >= 3.3.4, Follow Eigen Installation.

Opencv >= 3.2.0, FollowOpenCV_installation

Follow livox_ros_driver Installation.

Clone the repository and catkin_make:

    cd ~/$A_ROS_DIR$/src
    git clone https://github.com/YWL0720/FAST-LOCALIZATION
    cd FAST-LOCALIZATION
    git submodule update --init
    cd ../..
    catkin_make
    source devel/setup.bash

• Remember to source the livox_ros_driver before build (follow 1.3 livox_ros_driver)
• If you want to use a custom build of PCL, add the following line to ~/.bashrc export PCL_ROOT={CUSTOM_PCL_PATH}

FAST-LOCALIZATION can seamlessly integrate with HBA(A Globally Consistent and Efficient Large-Scale LiDAR Mapping Module), enabling real-time localization within maps optimized by HBA. It requires placing the point cloud file for each map frame and the corresponding pose files in the map/ directory. Noted the format of the pose is tx ty tz qw qx qy qz.

.
├── pcd
│   ├── 0.pcd
│   └── 1.pcd
└── pose.json

roslaunch fast_localization localization_mid360.launch

• Code Optimization
• Manual Initialization Interface

Thanks for FAST-LIO2, HBA and ScanContext