美文网首页ROS自主导航机器人
在树莓派4上编译运行orbslam2

在树莓派4上编译运行orbslam2

作者: PIBOT导航机器人 | 来源:发表于2020-10-27 10:40 被阅读0次

    1. 环境

    • 硬件环境:
      树莓派4B 4G 32G sandisk tf card
    • 软件环境:
    Linux pibot 5.4.0-1022-raspi #25~18.04.1-Ubuntu SMP PREEMPT Thu Oct 15 14:38:18 UTC 2020 aarch64 aarch64 aarch64 GNU/Linux
    pibot@pibot:~$ pkg-config --modversion opencv
    3.2.0
    pibot@pibot:~$ pkg-config --modversion eigen3
    3.3.4
    
    cd Pangolin
    git log --graph --decorate --oneline -1
    #* 86eb497 (HEAD -> master, origin/master, origin/HEAD) cmake: Fix wayland option
    
    cd ORB_SLAM2
    git log --graph --decorate --oneline -1
    # * f2e6f51 (HEAD -> master, origin/master, origin/HEAD) Update README.md
    

    2. 编译Pangolin

    sudo apt-get install libglew-dev
    unzip Pangolin.zip
    cd Pangolin
    mkdir build && cd build
    cmake ..
    make -j4
    sudo make install
    

    3. 编译orb-slam2

    3.1 修改头文件

    include/System.h新增#include <unistd.h>以修复usleep未定义的错误

    diff --git a/include/System.h b/include/System.h
    index b377b45..b2e990c 100644
    --- a/include/System.h
    +++ b/include/System.h
    @@ -25,7 +25,7 @@
     #include<string>
     #include<thread>
     #include<opencv2/core/core.hpp>
    +#include <unistd.h>
    
     #include "Tracking.h"
     #include "FrameDrawer.h"
     #include "MapDrawer.h"
    
    

    3.2 编译

    cd ORB_SLAM2
    git log --graph --decorate --oneline -1
    # * f2e6f51 (HEAD -> master, origin/master, origin/HEAD) Update README.md
    ./build.sh
    

    4. 测试

    4.1 数据集测试

    2.2 数据集准备

    cd dataset
    wget https://vision.in.tum.de/rgbd/dataset/freiburg1/rgbd_dataset_freiburg1_desk.tgz
    wget https://svncvpr.in.tum.de/cvpr-ros-pkg/trunk/rgbd_benchmark/rgbd_benchmark_tools/src/rgbd_benchmark_tools/associate.py
    tar xzvf rgbd_dataset_freiburg1_desk.tgz
    # 使用associate.py处理rgb.txt和depth.txt,生成配对的文件associate.txt
    cd rgbd_dataset_freiburg1_desk
    python ../associate.py rgb.txt depth.txt > associate.txt
    

    其他数据集下载地址

    4.2 测试

    cd ORB_SLAM2
    cd Vocabulary
    tar xzvf ORBvoc.txt.tar.gz # 解压得到ORBvoc.txt
    cd ..
    # Usage: ./rgbd_tum path_to_vocabulary path_to_settings path_to_sequence path_to_associatio
    ./Examples/RGB-D/rgbd_tum Vocabulary/ORBvoc.txt ./Examples/RGB-D/TUM1.yaml ../dataset/rgbd_dataset_freiburg1_desk ../dataset/rgbd_dataset_freiburg1_desk/associate.txt
    

    效果如下


    orb-slam2_freiburg1_desk

    相关文章

      网友评论

        本文标题:在树莓派4上编译运行orbslam2

        本文链接:https://www.haomeiwen.com/subject/gwwtvktx.html