xingyuuchen/lio-ppf

[IROS 2023] Fast LiDAR-Inertial Odometry via Incremental Plane Pre-Fitting and Skeleton Tracking

LIO-PPF

The official implementation of the paper "Fast LiDAR-Inertial Odometry via Incremental Plane Pre-Fitting and Skeleton Tracking" (IROS 2023)

We introduce LIO-PPF, a plane pre-fitting and skeleton tracking technique, that can ease the computation of state-of-the-art LIO systems, e.g. LIO-SAM. Please refer to this

In LIO-PPF, we track mainly the basic skeleton of the 3D scene, the planes of which are not fitted individually for each LiDAR scan, let alone for each LiDAR point. However, they are updated incrementally as the scene gradually `flows'.

By contrast, LIO-PPF can consume only 36% of the original local map size to achieve up to 4x faster residual computing and 1.92x overall FPS, while maintaining the same level of accuracy.

Quick Start

catkin_make
source devel/setup.bash
roslaunch lio_sam run.launch

In another terminal:

rosbag play /path/to/your/bag/file

For details about building and running, please refer to LIO-SAM.

FasterLIO with PPF

If you are looking for FasterLIO with PPF, please check out faster-lio-ppf.

Citation

If you find our work useful or interesting, please consider citing our paper:

@inproceedings{chen2023lio,
  title={LIO-PPF: Fast LiDAR-Inertial Odometry via Incremental Plane Pre-Fitting and Skeleton Tracking},
  author={Chen, Xingyu and
        Wu, Peixi and
        Li, Ge and
        Li, Thomas H},
  booktitle={2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={1458--1465},
  year={2023},
  organization={IEEE}
}

lio-ppf's Issues

关于在liosam使用nclt数据集

Hello, 首先感谢您们的工作！
我有一个关于您们如何得到 LIO-SAM 在nclt上的小问题, 如下:

我只把params.yaml三个地方改成了如下：
pointCloudTopic: "points_raw"
imuTopic: "imu_raw"
N_SCAN: 32
我想知道我是否漏掉了某个配置或改动，恳请指导，谢谢！

EU long term dataset

Hello, I saw that the [1] dataset was used in the paper, but I can't download it from the official channel now, can you share it?
[1] Z. Yan, L. Sun, T. Krajnik, and Y. Ruichek, “EU long-term dataset with multiple sensors for autonomous driving,” in Proceedings of the 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020.7

fast & faster lio with PPF

Hi @xingyuuchen Thanks for your great works and released code.
I see from your paper that you applied PPF to fast & faster lio， and achieved good results. fast & faster lio is more efficient than lio sam.
could you release the code of fast & faster lio with PPF. Thanks a lot.

open source

When it will open?

依赖glog

编译过程中，出现了以下错误，但我重新下载glog,gflags,还是没有解决，你遇到过这样的问题嘛

Kitti data

Hello author, thank you for your work, can you provide the configuration file for running on the kitti data set?

lio-ppf