• Stars
    star
    246
  • Rank 163,885 (Top 4 %)
  • Language
    C++
  • Created almost 4 years ago
  • Updated almost 3 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Lvio-Fusion: A Self-adaptive Multi-sensor Fusion SLAM Framework Using Actor-critic Method (IROS 2021)

lvio_fusion

Introduction

A Self-adaptive Multi-sensor Fusion SLAM Framework Using Actor-critic Method, which focus on the complex environment of vehicles.

Here is the paper.

@misc{jia2021lviofusion,
      title={Lvio-Fusion: A Self-adaptive Multi-sensor Fusion SLAM Framework Using Actor-critic Method}, 
      author={Yupeng Jia and Haiyong Luo and Fang Zhao and Guanlin Jiang and Yuhang Li and Jiaquan Yan and Zhuqing Jiang},
      year={2021},
      eprint={2106.06783},
      archivePrefix={arXiv},
      primaryClass={cs.RO}
}

Dependencies

  • ros (Kinetic/Melodic/Noetic)
  • Eigen3
  • Sophus
  • opencv
  • pcl
  • ceres-solver
  • libgeographic-dev

Usage

Complie:

catkin_make

Run:

source devel/setup.bash
roslaunch lvio_fusion_node kitti.launch

Result

kitti: To generate rosbags using KITTI raw data, you can use the script in lio-sam and the configuration file kitti2.yaml

kaist urban: Here is a video: video