FAST_LIO_LOCALIZATION重定位模块，在ubuntu18、ros melodic、python2.7、livox ros driver2环境下安装流程及相关报错解决

Sanse_

已于 2024-07-30 09:39:52 修改

阅读量1.9k

点赞数 27

分类专栏： Linux、ros知识编译、代码运行算法相关文章标签：算法 ubuntu 自动驾驶

于 2024-07-30 09:33:11 首次发布

本文链接：https://blog.csdn.net/Sanse_/article/details/140788151

版权

编译、代码运行同时被 3 个专栏收录

5 篇文章

订阅专栏

Linux、ros知识

4 篇文章

订阅专栏

算法相关

3 篇文章

订阅专栏

首先是官方的下载地址和安装参考，列举如下

GitHub - HViktorTsoi/FAST_LIO_LOCALIZATION: A simple localization framework that can re-localize in built maps based on FAST-LIO.

2.1 Dependencies for FAST-LIO

Technically, if you have built and run FAST-LIO before, you may skip section 2.1.

This part of dependency is consistent with FAST-LIO, please refer to the documentation GitHub - hku-mars/FAST_LIO: A computationally efficient and robust LiDAR-inertial odometry (LIO) package

2.2 Dependencies for localization module

python 2.7
ros_numpy

sudo apt install ros-$ROS_DISTRO-ros-numpy

Open3D

pip install open3d==0.9

Notice that, there may be issue when installing Open3D directly using pip in Python2.7:

ERROR: Package 'pyrsistent' requires a different Python: 2.7.18 not in'>=3.5'

you may firstly install pyrsistent:

pip install pyrsistent==0.15

Then

pip install open3d==0.9

3. Build

Clone the repository and catkin_make:

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

Remember to source the livox_ros_driver before build (follow 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}

其环境配置是要求python2.7版本、pcl库的话我看有些地方写需要pcl1.8，但是我安装的pcl1.7.2是同样能跑通的、以及rosnumpy、open3d。注意这里的rosnumpy和open3d都需要在python2.7的环境下进行安装，因为有的人linux下安装了多个python版本，有2.7、3.5、3.6等版本，如果你在python3.6下安装了open3d，你在运行代码的时候一定会报错找不到open3d模块的，这里一定要注意，一定要在python2.7环境下安装。

自己调整的完整安装过程

跑通FAST_LIO_LOCALIZATION记录_fastlio localization-CSDN博客

conda create -n fastliolocationpy27 python=2.7
conda activate fastliolocationpy27
pip install -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com open3d==0.9
pip install rospy

Looking in indexes: http://pypi.douban.com/simple/

报错：ERROR: Could not find a version that satisfies the requirement open3d==0.9 (from versions: none)

ERROR: No matching distribution found for open3d==0.9

这是因为指定了数据源，这个豆瓣的数据源不管用了，没有这个对应版本

报错：ERROR: Could not find a version that satisfies the requirement comm>=0.1.3 (from ipywidgets) (from versions: 0.0.1)

ERROR: No matching distribution found for comm>=0.1.3 (from ipywidgets)

这是因为自带的 ipywidgets下载新版，所产生的comm的版本不正确，因此指定一个老版本的ipywidgets安装，解决问题

pip install ipywidgets==7.6.3
pip install open3d

继续安装

pip install  open3d==0.9
pip install rospy

因为安装了conda环境重新制定了python解释器的位置，所以需要把global_localization.py的第一行改成虚拟环境下python2.7的路径

#!/home/zeratul/anaconda3/envs/fastliolocationpy27/bin/python2.7

在对应终端中继续执行以下指令

#激活conda环境，在这个环境之下安装其他的包，才不会出错   
conda activate fastliolocationpy27
#跳转到工作空间src目录，拉取代码
cd ~/$A_ROS_DIR$/src    
git clone https://github.com/HViktorTsoi/FAST_LIO_LOCALIZATION.git
cd FAST_LIO_LOCALIZATION
git submodule update --init
cd ../..
#找不到“livox_ros_driver2”提供的包配置文件，这是因为没有安装或者在安装livox_ros_driver2之后没有使用source ./devel/setup.sh命令更新当前ROS包环境。注意这里在代码里面更新为了2代
source /home/zeratul/project/FAST-LIO2/devel/setup.bash
#进行编译，这里容易出错    
catkin_make
#启动前source一下
source devel/setup.bash

注意：source ./devel/setup.sh 该命令只在当前的终端中有效，所以报错有可能是因为你在新建的终端窗口中运行 catkin_make。这种情况只需要先去指定目录下先source一下再回到～/catkin_ws/src目录下catkin_make，即可解决。

修改livox_ros_driver的版本来适配代码

在cmakelist、package.xml、整体的代码中，查找livox_ros_driver去替换为livox_ros_driver2

报错fast_lio/Pose6D.h，没有那个文件或目录

这个问题我这里也存疑，有懂的大佬可以详解一下，可能和安装方式有关。

网上的解决一般有两种：

有些地方解答说是需要进入FAST_LIO_LOCALIZATION/include/common_lib.h文件中，把这个文件中的所有fast_lio都改为fast_lio_localization。
原文链接： FAST_LIO_LOCALIZATION的安装编译_fastlio安装-CSDN博客
后就是cmakelist中的项目名称,统一改成fast_lio,因为不改的话,后面编译的时候就会找不到pose6d这个自定义消息,#include <fast_lio/Pose6D.h> 因为是这样引用的.
原文链接：基于fastlio的重定位过程_fast lio定位-CSDN博客

这是两个不同思路，因为FAST_LIO_LOCALIZATION里面也有Pose6D.msg，可以生成.h文件，结果就是引用Posd6D.h文件属于FAST_LIO或者FAST_LIO_LOCALIZATION的区别。

我起初安装各种报错，然后就新建了工作空间，专门pull下来了FAST_LIO_LOCALIZATION的代码，然后进行编译，报了这个错误。按照第一种方式进行修改，仍然产生报错：fast_lio_localization/Pose6D.h，没有那个文件或目录。查阅了一下资料，代码中有一个Pose6D.msg并没有生成对应.h，有回答如下：

猜测：生成不了pose6D的原因，主要在于找的依赖不对，依赖不全，不能够从pose6d.msg生成对应的.h

官方代码安装第一部分这么说：

2.1 Dependencies for FAST-LIO
Technically, if you have built and run FAST-LIO before, you may skip section 2.1.
This part of dependency is consistent with FAST-LIO, please refer to the documentation GitHub - hku-mars/FAST_LIO: A computationally efficient and robust LiDAR-inertial odometry (LIO) package

有一点不甚明确，如果单独编译FAST_LIO_LOCALIZATION，可能需要在src下增加一定的依赖一块编译，如livox_ros_driver2，单独source这个代码可能不行。没有仔细探究，直接把代码pull到了FAST_LIO和livox_ros_driver2共同编译好的工作空间，再编译一次。

其他可能得解决方案：

Pose6D.h problem during catkin_make · Issue #138 · hku-mars/FAST_LIO · GitHub

After i build it for the second time i had same error, i have found discussions about creating custom ROS messages and found difference between their CMakeLists and this repo's. I have just added genmsg into find_package
find_package(catkin REQUIRED COMPONENTS geometry_msgs nav_msgs sensor_msgs roscpp rospy std_msgs pcl_ros tf livox_ros_driver message_generation eigen_conversions genmsg )
and also added new line
add_dependencies(fastlio_mapping fast_lio_generate_messages_cpp)
That worked for me

Please check your running environment, it might be caused by the hardware
I experienced same error message, and I found this case caused by limited CPU cores and RAM.
I used running this process in a VirtualBox machine with only 1 CPU and 2GB RAM, after I increased CPU cores to 4 cores and 4GB RAM, then problem resolved.
CMakeFiles/Makefile2:531: recipe for target 'FAST_LIO/CMakeFiles/fastlio_mapping.dir/all' failed
Upgrading RAM from 4GB to 8GB solved issue for Ubuntu 18.04, ROS Melodic on VMware Fusion.

try catkin_make -j1

I have the same problem of Pose6D.h during catkin_make of FAST-LIO2. My environment is Ubuntu 20 with ROS noetic. I'm using livox_ros_driver2 and livox_sdk2 for mid-360 lidar. I tried using catkin_make -j1 but the build problem still existed.

I realised that I put FAST-LIO under the same workspace as the livox_ros_driver2. So I managed to solve the problem in this way

Separate the workspace for livox_ros_driver2 as the ros driver is built using its own ./build.sh ROS1 script. Source the livox_ros_driver2 workspace.
Put the FAST-LIO and POINT-LIO in another workspace folder and catkin_make as usual.
I hope this helps other people having the same problem.

报错：add_executable cannot create target "fastlio_mapping" because another target because another target with the same name

说明有两个名称相同的文件同时经过了可执行设置，有冲突。此时检查一下两个重名文件内容是否相同。若不相同考虑改名，但同时也要改变CMakeLists.txt配置中所有对应的文件名称。若相同，不要删除任何一个文件，将不需要的文件的可执行设置及其他所有配置在其所在功能包的CMakeLists.txt中注释掉或者改掉即可。

类似下面：

参考了这里的解释：

在配置时，我们发现了一些问题，主要还是版本导致。
一个是原仓库使用的是python2，python2使用和安装外部库已经不太方便了，所以换成了python3。
另一个是FAST_LIO_LOCALIZATION里面也包含了FAST_LIO，但是这里面的FAST_LIO还是比较老的版本，建议还是从最新的FAST_LIO仓库里拉最新的(因为做了mid360的适配)，然后记得在 cmakelist里修改生成的可执行文件的名称，否则会和上一步的fast_lio的mapping重复报错。
————————————————
原文链接：使用mid360从0开始搭建实物机器人入门级导航系统，基于Fast_Lio,Move_Base-CSDN博客

也就是说，当一个工作空间中，同时存在FAST_LIO_LOCALIZATION与FAST_LIO时，由于都含有fastlio_mapping模块，生成可执行文件与连接的过程，系统不知道该链接给那一块的fastlio_mapping了。也就是这里重复了：