ubantu18.04：ORB_SlAM3安装 从零到一 包学会！

最新推荐文章于 2025-05-07 09:54:54 发布
最新推荐文章于 2025-05-07 09:54:54 发布
阅读量841 收藏 8
点赞数 11
文章标签： 自动驾驶 人工智能 深度学习 机器人
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_47006778/article/details/147747823
版权

前言

系统环境：ubantu18.04 虚拟机

提前安装:

wget http://fishros.com/install -O fishros && . fishros

sudo apt update
sudo apt install git cmake gcc g++

断网解决办法：

nmcli networking on

参考：

https://blog.csdn.net/weixin_48924581/article/details/126534339

https://blog.csdn.net/shadowmimii/article/details/129472719

https://blog.csdn.net/Zoey_chn/article/details/135139494

ORB-SLAM3与ROS集成

ORB-SLAM3可以与ROS（Robot Operating System）集成，以支持更多的机器人应用。以下是一些相关的ROS包：

  • orb_slam3_ros：一个将ORB-SLAM3集成到ROS中的包，支持单目、双目和RGB-D SLAM。
  • orb_slam3_ros_wrapper：一个更高级的ROS包装器，提供了更多的功能和配置选项。

相关工具和库

  • DBoW2：用于图像序列中的快速地点识别。
  • g2o：一个通用的图优化库，用于SLAM中的地图优化。
  • Eigen：一个C++模板库，用于线性代数运算。

详解：ORB-SLAM3代码详解----基本框架解析及ORB特征提取_orbslam3代码详解-CSDN博客

1、安装EIGEN3.3.4

直接apt安装

sudo apt-get install libeigen3-dev

2、安装pangolin0.6

sudo apt install libglew-dev libpython2.7-dev

https://github.com/stevenlovegrove/Pangolin/releases/tag/v0.6

获得pangolin的稳定版本安装包后进行编译，

mkdir build
cd build
cmake ..
make
sudo make install	
//安装后pangolin将在  /usr/local/include/中找到

详细编译记录：

ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6$ mkdir build
ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6$ cd build
ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$ cmake ..
-- The C compiler identification is GNU 7.5.0
-- The CXX compiler identification is GNU 7.5.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
Build type not set (defaults to release)
-DCMAKE_BUILD_TYPE=Debug for debug
-- Found OpenGL: /usr/lib/x86_64-linux-gnu/libOpenGL.so   
-- Found GLEW: /usr/lib/x86_64-linux-gnu/libGLEW.so
-- Eigen Found and Enabled
-- libdc1394 Found and Enabled
-- V4L Found and Enabled
-- Performing Test HAVE_FFMPEG_MAX_ANALYZE_DURATION
-- Performing Test HAVE_FFMPEG_MAX_ANALYZE_DURATION - Success
-- Performing Test HAVE_FFMPEG_AVFORMAT_ALLOC_OUTPUT_CONTEXT2
-- Performing Test HAVE_FFMPEG_AVFORMAT_ALLOC_OUTPUT_CONTEXT2 - Success
-- Performing Test HAVE_FFMPEG_AVPIXELFORMAT
-- Performing Test HAVE_FFMPEG_AVPIXELFORMAT - Success
-- ffmpeg Found and Enabled
-- Found PkgConfig: /usr/bin/pkg-config (found version "0.29.1") 
-- OpenNI Found and Enabled
-- OpenNI2 Found and Enabled
-- libpng Found and Enabled
-- libjpeg Found and Enabled
-- libtiff Found and Enabled
-- libopenexr Found and Enabled
-- liblz4 Found and Enabled
-- Could NOT find Doxygen (missing: DOXYGEN_EXECUTABLE) 
-- Configuring done
-- Generating done
-- Build files have been written to: /home/ht_llibra/VSlam/Pangolin-v0.6/build
ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$ 


ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$ make
Scanning dependencies of target pangolin
[  1%] Building CXX object src/CMakeFiles/pangolin.dir/utils/file_extension.cpp.o
[  2%] Building CXX object src/CMakeFiles/pangolin.dir/utils/file_utils.cpp.o
[  2%] Building CXX object src/CMakeFiles/pangolin.dir/utils/sigstate.cpp.o
[  3%] Building CXX object src/CMakeFiles/pangolin.dir/utils/threadedfilebuf.cpp.o
[  4%] Building CXX object src/CMakeFiles/pangolin.dir/utils/timer.cpp.o
[  5%] Building CXX object src/CMakeFiles/pangolin.dir/utils/uri.cpp.o
[  6%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io.cpp.o
[  7%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_exr.cpp.o
[  7%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_jpg.cpp.o
[  8%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_lz4.cpp.o
[  9%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_packed12bit.cpp.o
[ 10%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_pango.cpp.o
[ 11%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_png.cpp.o
[ 12%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_ppm.cpp.o
[ 12%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_raw.cpp.o
[ 13%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_tga.cpp.o
[ 14%] Building CXX object src/CMakeFiles/pangolin.dir/image/image_io_zstd.cpp.o
[ 15%] Building CXX object src/CMakeFiles/pangolin.dir/image/pixel_format.cpp.o
[ 16%] Building CXX object src/CMakeFiles/pangolin.dir/log/packet.cpp.o
[ 17%] Building CXX object src/CMakeFiles/pangolin.dir/log/packetstream.cpp.o
[ 17%] Building CXX object src/CMakeFiles/pangolin.dir/log/packetstream_reader.cpp.o
[ 18%] Building CXX object src/CMakeFiles/pangolin.dir/log/packetstream_writer.cpp.o
[ 19%] Building CXX object src/CMakeFiles/pangolin.dir/log/playback_session.cpp.o
[ 20%] Building CXX object src/CMakeFiles/pangolin.dir/geometry/geometry.cpp.o
[ 21%] Building CXX object src/CMakeFiles/pangolin.dir/geometry/geometry_obj.cpp.o
[ 22%] Building CXX object src/CMakeFiles/pangolin.dir/geometry/geometry_ply.cpp.o
[ 22%] Building CXX object src/CMakeFiles/pangolin.dir/geometry/glgeometry.cpp.o
[ 23%] Building CXX object src/CMakeFiles/pangolin.dir/geometry/tinyobj.cpp.o
[ 24%] Building CXX object src/CMakeFiles/pangolin.dir/gl/glchar.cpp.o
[ 25%] Building CXX object src/CMakeFiles/pangolin.dir/gl/gldraw.cpp.o
[ 26%] Building CXX object src/CMakeFiles/pangolin.dir/gl/glfont.cpp.o
[ 27%] Building CXX object src/CMakeFiles/pangolin.dir/gl/glpangoglu.cpp.o
[ 27%] Building CXX object src/CMakeFiles/pangolin.dir/gl/gltext.cpp.o
[ 28%] Building CXX object src/CMakeFiles/pangolin.dir/gl/gltexturecache.cpp.o
[ 29%] Building CXX object src/CMakeFiles/pangolin.dir/display/display.cpp.o
[ 30%] Building CXX object src/CMakeFiles/pangolin.dir/display/image_view.cpp.o
[ 31%] Building CXX object src/CMakeFiles/pangolin.dir/display/opengl_render_state.cpp.o
[ 32%] Building CXX object src/CMakeFiles/pangolin.dir/display/view.cpp.o
[ 32%] Building CXX object src/CMakeFiles/pangolin.dir/display/viewport.cpp.o
[ 33%] Building CXX object src/CMakeFiles/pangolin.dir/display/window_factory.cpp.o
[ 34%] Building CXX object src/CMakeFiles/pangolin.dir/handler/handler.cpp.o
[ 35%] Building CXX object src/CMakeFiles/pangolin.dir/handler/handler_glbuffer.cpp.o
[ 36%] Building CXX object src/CMakeFiles/pangolin.dir/handler/handler_image.cpp.o
[ 37%] Building CXX object src/CMakeFiles/pangolin.dir/plot/datalog.cpp.o
[ 37%] Building CXX object src/CMakeFiles/pangolin.dir/plot/plotter.cpp.o
[ 38%] Building CXX object src/CMakeFiles/pangolin.dir/var/input_record_repeat.cpp.o
[ 39%] Building CXX object src/CMakeFiles/pangolin.dir/var/vars.cpp.o
[ 40%] Building CXX object src/CMakeFiles/pangolin.dir/display/widgets/widgets.cpp.o
[ 41%] Building CXX object src/CMakeFiles/pangolin.dir/video/stream_encoder_factory.cpp.o
[ 42%] Building CXX object src/CMakeFiles/pangolin.dir/video/video.cpp.o
[ 42%] Building CXX object src/CMakeFiles/pangolin.dir/video/video_input.cpp.o
[ 43%] Building CXX object src/CMakeFiles/pangolin.dir/video/video_interface_factory.cpp.o
[ 44%] Building CXX object src/CMakeFiles/pangolin.dir/video/video_output.cpp.o
[ 45%] Building CXX object src/CMakeFiles/pangolin.dir/video/video_output_interface_factory.cpp.o
[ 46%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/test.cpp.o
[ 47%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/images.cpp.o
[ 47%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/images_out.cpp.o
[ 48%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/split.cpp.o
[ 49%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/truncate.cpp.o
[ 50%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/pvn.cpp.o
[ 51%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/pango.cpp.o
[ 52%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/pango_video_output.cpp.o
[ 52%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/debayer.cpp.o
[ 53%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/shift.cpp.o
[ 54%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/mirror.cpp.o
[ 55%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/unpack.cpp.o
[ 56%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/pack.cpp.o
[ 57%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/join.cpp.o
[ 57%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/merge.cpp.o
[ 58%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/json.cpp.o
[ 59%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/thread.cpp.o
[ 60%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/shared_memory.cpp.o
[ 61%] Building CXX object src/CMakeFiles/pangolin.dir/tools/video_viewer.cpp.o
[ 62%] Building CXX object src/CMakeFiles/pangolin.dir/display/device/display_headless.cpp.o
[ 62%] Building CXX object src/CMakeFiles/pangolin.dir/display/device/display_x11.cpp.o
[ 63%] Building CXX object src/CMakeFiles/pangolin.dir/utils/posix/condition_variable.cpp.o
[ 64%] Building CXX object src/CMakeFiles/pangolin.dir/utils/posix/semaphore.cpp.o
[ 65%] Building CXX object src/CMakeFiles/pangolin.dir/utils/posix/shared_memory_buffer.cpp.o
[ 66%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/firewire.cpp.o
[ 67%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/deinterlace.cpp.o
[ 67%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/v4l.cpp.o
[ 68%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/ffmpeg.cpp.o
[ 69%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/openni.cpp.o
/home/ht_llibra/VSlam/Pangolin-v0.6/src/video/drivers/openni.cpp: In constructor ‘pangolin::OpenNiVideo::OpenNiVideo(pangolin::OpenNiSensorType, pangolin::OpenNiSensorType, pangolin::ImageDim, int)’:
/home/ht_llibra/VSlam/Pangolin-v0.6/src/video/drivers/openni.cpp:82:28: warning: this statement may fall through [-Wimplicit-fallthrough=]
             depth_to_color = true;
             ~~~~~~~~~~~~~~~^~~~~~
/home/ht_llibra/VSlam/Pangolin-v0.6/src/video/drivers/openni.cpp:83:9: note: here
         case OpenNiDepth_1mm:
         ^~~~
[ 70%] Building CXX object src/CMakeFiles/pangolin.dir/video/drivers/openni2.cpp.o
[ 71%] Building CXX object src/CMakeFiles/pangolin.dir/fonts.cpp.o
[ 72%] Linking CXX shared library libpangolin.so
[ 72%] Built target pangolin
Scanning dependencies of target VideoViewer
[ 73%] Building CXX object tools/VideoViewer/CMakeFiles/VideoViewer.dir/main.cpp.o
[ 74%] Linking CXX executable VideoViewer
[ 74%] Built target VideoViewer
Scanning dependencies of target VideoConvert
[ 75%] Building CXX object tools/VideoConvert/CMakeFiles/VideoConvert.dir/main.cpp.o
[ 76%] Linking CXX executable VideoConvert
[ 76%] Built target VideoConvert
Scanning dependencies of target VideoJsonPrint
[ 77%] Building CXX object tools/VideoJson/CMakeFiles/VideoJsonPrint.dir/main-print.cpp.o
[ 78%] Linking CXX executable VideoJsonPrint
[ 78%] Built target VideoJsonPrint
Scanning dependencies of target VideoJsonTransform
[ 78%] Building CXX object tools/VideoJson/CMakeFiles/VideoJsonTransform.dir/main-transform.cpp.o
[ 79%] Linking CXX executable VideoJsonTransform
[ 79%] Built target VideoJsonTransform
Scanning dependencies of target Plotter
[ 80%] Building CXX object tools/Plotter/CMakeFiles/Plotter.dir/main.cpp.o
[ 81%] Linking CXX executable Plotter
[ 81%] Built target Plotter
Scanning dependencies of target ModelViewer
[ 81%] Building CXX object tools/ModelViewer/CMakeFiles/ModelViewer.dir/main.cpp.o
[ 82%] Linking CXX executable ModelViewer
[ 82%] Built target ModelViewer
Scanning dependencies of target HelloPangolin
[ 82%] Building CXX object examples/HelloPangolin/CMakeFiles/HelloPangolin.dir/main.cpp.o
[ 83%] Linking CXX executable HelloPangolin
[ 83%] Built target HelloPangolin
Scanning dependencies of target HelloPangolinOffscreen
[ 84%] Building CXX object examples/HelloPangolinOffscreen/CMakeFiles/HelloPangolinOffscreen.dir/main.cpp.o
[ 85%] Linking CXX executable HelloPangolinOffscreen
[ 85%] Built target HelloPangolinOffscreen
Scanning dependencies of target HelloPangolinThreads
[ 86%] Building CXX object examples/HelloPangolinThreads/CMakeFiles/HelloPangolinThreads.dir/main.cpp.o
[ 87%] Linking CXX executable HelloPangolinThreads
[ 87%] Built target HelloPangolinThreads
Scanning dependencies of target SimpleMultiDisplay
[ 88%] Building CXX object examples/SimpleMultiDisplay/CMakeFiles/SimpleMultiDisplay.dir/main.cpp.o
[ 89%] Linking CXX executable SimpleMultiDisplay
[ 89%] Built target SimpleMultiDisplay
Scanning dependencies of target SimpleDisplayImage
[ 90%] Building CXX object examples/SimpleDisplayImage/CMakeFiles/SimpleDisplayImage.dir/main.cpp.o
[ 91%] Linking CXX executable SimpleDisplayImage
[ 91%] Built target SimpleDisplayImage
Scanning dependencies of target SimpleScene
[ 92%] Building CXX object examples/SimpleScene/CMakeFiles/SimpleScene.dir/main.cpp.o
[ 93%] Linking CXX executable SimpleScene
[ 93%] Built target SimpleScene
Scanning dependencies of target SimplePlot
[ 93%] Building CXX object examples/SimplePlot/CMakeFiles/SimplePlot.dir/main.cpp.o
[ 94%] Linking CXX executable SimplePlot
[ 94%] Built target SimplePlot
Scanning dependencies of target SimpleDisplay
[ 94%] Building CXX object examples/SimpleDisplay/CMakeFiles/SimpleDisplay.dir/main.cpp.o
[ 95%] Linking CXX executable SimpleDisplay
[ 95%] Built target SimpleDisplay
Scanning dependencies of target SimpleVideo
[ 95%] Building CXX object examples/SimpleVideo/CMakeFiles/SimpleVideo.dir/main.cpp.o
[ 96%] Linking CXX executable SimpleVideo
[ 96%] Built target SimpleVideo
Scanning dependencies of target SimpleRecord
[ 97%] Building CXX object examples/SimpleRecord/CMakeFiles/SimpleRecord.dir/main.cpp.o
[ 98%] Linking CXX executable SimpleRecord
[ 98%] Built target SimpleRecord
Scanning dependencies of target SharedMemoryCamera
[ 99%] Building CXX object examples/SharedMemoryCamera/CMakeFiles/SharedMemoryCamera.dir/main.cpp.o
[100%] Linking CXX executable SharedMemoryCamera
[100%] Built target SharedMemoryCamera
ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$ 


ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$ sudo make install
[sudo] ht_llibra 的密码： 
[ 72%] Built target pangolin
[ 74%] Built target VideoViewer
[ 76%] Built target VideoConvert
[ 78%] Built target VideoJsonPrint
[ 79%] Built target VideoJsonTransform
[ 81%] Built target Plotter
[ 82%] Built target ModelViewer
[ 83%] Built target HelloPangolin
[ 85%] Built target HelloPangolinOffscreen
[ 87%] Built target HelloPangolinThreads
[ 89%] Built target SimpleMultiDisplay
[ 91%] Built target SimpleDisplayImage
[ 93%] Built target SimpleScene
[ 94%] Built target SimplePlot
[ 95%] Built target SimpleDisplay
[ 96%] Built target SimpleVideo
[ 98%] Built target SimpleRecord
[100%] Built target SharedMemoryCamera
Install the project...
-- Install configuration: "Release"
-- Installing: /usr/local/include/pangolin/config.h
-- Up-to-date: /usr/local/include/pangolin
-- Installing: /usr/local/include/pangolin/compat
-- Installing: /usr/local/include/pangolin/compat/glutbitmap.h
-- Installing: /usr/local/include/pangolin/compat/variant.h
-- Installing: /usr/local/include/pangolin/compat/optional.h
-- Installing: /usr/local/include/pangolin/compat/type_traits.h
-- Installing: /usr/local/include/pangolin/console
-- Installing: /usr/local/include/pangolin/console/ConsoleView.h
-- Installing: /usr/local/include/pangolin/console/ConsoleInterpreter.h
-- Installing: /usr/local/include/pangolin/var
-- Installing: /usr/local/include/pangolin/var/varvaluet.h
-- Installing: /usr/local/include/pangolin/var/varstate.h
-- Installing: /usr/local/include/pangolin/var/varwrapper.h
-- Installing: /usr/local/include/pangolin/var/var.h
-- Installing: /usr/local/include/pangolin/var/input_record_repeat.h
-- Installing: /usr/local/include/pangolin/var/varvalue.h
-- Installing: /usr/local/include/pangolin/var/varextra.h
-- Installing: /usr/local/include/pangolin/var/varvaluegeneric.h
-- Installing: /usr/local/include/pangolin/log
-- Installing: /usr/local/include/pangolin/log/sync_time.h
-- Installing: /usr/local/include/pangolin/log/packetstream.h
-- Installing: /usr/local/include/pangolin/log/playback_session.h
-- Installing: /usr/local/include/pangolin/log/packet.h
-- Installing: /usr/local/include/pangolin/log/packetstream_tags.h
-- Installing: /usr/local/include/pangolin/log/packetstream_source.h
-- Installing: /usr/local/include/pangolin/log/packetstream_reader.h
-- Installing: /usr/local/include/pangolin/log/packetstream_writer.h
-- Installing: /usr/local/include/pangolin/gl
-- Installing: /usr/local/include/pangolin/gl/gltext.h
-- Installing: /usr/local/include/pangolin/gl/compat
-- Installing: /usr/local/include/pangolin/gl/compat/gl_es_compat.h
-- Installing: /usr/local/include/pangolin/gl/compat/gl2engine.h
-- Installing: /usr/local/include/pangolin/gl/glsl.h
-- Installing: /usr/local/include/pangolin/gl/glchar.h
-- Installing: /usr/local/include/pangolin/gl/gldraw.h
-- Installing: /usr/local/include/pangolin/gl/glinclude.h
-- Installing: /usr/local/include/pangolin/gl/glplatform.h
-- Installing: /usr/local/include/pangolin/gl/glstate.h
-- Installing: /usr/local/include/pangolin/gl/cg.h
-- Installing: /usr/local/include/pangolin/gl/glvbo.h
-- Installing: /usr/local/include/pangolin/gl/glfont.h
-- Installing: /usr/local/include/pangolin/gl/gltexturecache.h
-- Installing: /usr/local/include/pangolin/gl/glcuda.h
-- Installing: /usr/local/include/pangolin/gl/glpixformat.h
-- Installing: /usr/local/include/pangolin/gl/gl.h
-- Installing: /usr/local/include/pangolin/gl/glformattraits.h
-- Installing: /usr/local/include/pangolin/gl/gl.hpp
-- Installing: /usr/local/include/pangolin/gl/colour.h
-- Installing: /usr/local/include/pangolin/gl/glpangoglu.h
-- Installing: /usr/local/include/pangolin/scene
-- Installing: /usr/local/include/pangolin/scene/renderable.h
-- Installing: /usr/local/include/pangolin/scene/scenehandler.h
-- Installing: /usr/local/include/pangolin/scene/axis.h
-- Installing: /usr/local/include/pangolin/scene/interactive_index.h
-- Installing: /usr/local/include/pangolin/scene/tree.h
-- Installing: /usr/local/include/pangolin/scene/interactive.h
-- Installing: /usr/local/include/pangolin/tools
-- Installing: /usr/local/include/pangolin/tools/video_viewer.h
-- Installing: /usr/local/include/pangolin/image
-- Installing: /usr/local/include/pangolin/image/image_io.h
-- Installing: /usr/local/include/pangolin/image/image_convert.h
-- Installing: /usr/local/include/pangolin/image/image.h
-- Installing: /usr/local/include/pangolin/image/image_utils.h
-- Installing: /usr/local/include/pangolin/image/memcpy.h
-- Installing: /usr/local/include/pangolin/image/typed_image.h
-- Installing: /usr/local/include/pangolin/image/managed_image.h
-- Installing: /usr/local/include/pangolin/image/copy.h
-- Installing: /usr/local/include/pangolin/image/pixel_format.h
-- Installing: /usr/local/include/pangolin/video
-- Installing: /usr/local/include/pangolin/video/drivers
-- Installing: /usr/local/include/pangolin/video/drivers/merge.h
-- Installing: /usr/local/include/pangolin/video/drivers/thread.h
-- Installing: /usr/local/include/pangolin/video/drivers/unpack.h
-- Installing: /usr/local/include/pangolin/video/drivers/pvn.h
-- Installing: /usr/local/include/pangolin/video/drivers/shared_memory.h
-- Installing: /usr/local/include/pangolin/video/drivers/realsense.h
-- Installing: /usr/local/include/pangolin/video/drivers/shift.h
-- Installing: /usr/local/include/pangolin/video/drivers/openni.h
-- Installing: /usr/local/include/pangolin/video/drivers/split.h
-- Installing: /usr/local/include/pangolin/video/drivers/openni2.h
-- Installing: /usr/local/include/pangolin/video/drivers/images.h
-- Installing: /usr/local/include/pangolin/video/drivers/join.h
-- Installing: /usr/local/include/pangolin/video/drivers/v4l.h
-- Installing: /usr/local/include/pangolin/video/drivers/uvc_mediafoundation.h
-- Installing: /usr/local/include/pangolin/video/drivers/pango.h
-- Installing: /usr/local/include/pangolin/video/drivers/truncate.h
-- Installing: /usr/local/include/pangolin/video/drivers/realsense2.h
-- Installing: /usr/local/include/pangolin/video/drivers/uvc.h
-- Installing: /usr/local/include/pangolin/video/drivers/pack.h
-- Installing: /usr/local/include/pangolin/video/drivers/firewire.h
-- Installing: /usr/local/include/pangolin/video/drivers/test.h
-- Installing: /usr/local/include/pangolin/video/drivers/deinterlace.h
-- Installing: /usr/local/include/pangolin/video/drivers/debayer.h
-- Installing: /usr/local/include/pangolin/video/drivers/teli.h
-- Installing: /usr/local/include/pangolin/video/drivers/depthsense.h
-- Installing: /usr/local/include/pangolin/video/drivers/openni_common.h
-- Installing: /usr/local/include/pangolin/video/drivers/ffmpeg.h
-- Installing: /usr/local/include/pangolin/video/drivers/mirror.h
-- Installing: /usr/local/include/pangolin/video/drivers/images_out.h
-- Installing: /usr/local/include/pangolin/video/drivers/pango_video_output.h
-- Installing: /usr/local/include/pangolin/video/drivers/pleora.h
-- Installing: /usr/local/include/pangolin/video/stream_info.h
-- Installing: /usr/local/include/pangolin/video/video_input.h
-- Installing: /usr/local/include/pangolin/video/video_output_interface.h
-- Installing: /usr/local/include/pangolin/video/video_exception.h
-- Installing: /usr/local/include/pangolin/video/video.h
-- Installing: /usr/local/include/pangolin/video/stream_encoder_factory.h
-- Installing: /usr/local/include/pangolin/video/iostream_operators.h
-- Installing: /usr/local/include/pangolin/video/video_output.h
-- Installing: /usr/local/include/pangolin/video/video_record_repeat.h
-- Installing: /usr/local/include/pangolin/video/video_interface.h
-- Installing: /usr/local/include/pangolin/platform.h
-- Installing: /usr/local/include/pangolin/handler
-- Installing: /usr/local/include/pangolin/handler/handler_image.h
-- Installing: /usr/local/include/pangolin/handler/handler.h
-- Installing: /usr/local/include/pangolin/handler/handler_glbuffer.h
-- Installing: /usr/local/include/pangolin/handler/handler_enums.h
-- Installing: /usr/local/include/pangolin/utils
-- Installing: /usr/local/include/pangolin/utils/log.h
-- Installing: /usr/local/include/pangolin/utils/format_string.h
-- Installing: /usr/local/include/pangolin/utils/signal_slot.h
-- Installing: /usr/local/include/pangolin/utils/params.h
-- Installing: /usr/local/include/pangolin/utils/file_utils.h
-- Installing: /usr/local/include/pangolin/utils/argagg.hpp
-- Installing: /usr/local/include/pangolin/utils/parse.h
-- Installing: /usr/local/include/pangolin/utils/threadedfilebuf.h
-- Installing: /usr/local/include/pangolin/utils/variadic_all.h
-- Installing: /usr/local/include/pangolin/utils/file_extension.h
-- Installing: /usr/local/include/pangolin/utils/type_convert.h
-- Installing: /usr/local/include/pangolin/utils/sigstate.h
-- Installing: /usr/local/include/pangolin/utils/simple_math.h
-- Installing: /usr/local/include/pangolin/utils/timer.h
-- Installing: /usr/local/include/pangolin/utils/fix_size_buffer_queue.h
-- Installing: /usr/local/include/pangolin/utils/transform.h
-- Installing: /usr/local/include/pangolin/utils/picojson.h
-- Installing: /usr/local/include/pangolin/utils/assert.h
-- Installing: /usr/local/include/pangolin/utils/memstreambuf.h
-- Installing: /usr/local/include/pangolin/utils/posix
-- Installing: /usr/local/include/pangolin/utils/posix/condition_variable.h
-- Installing: /usr/local/include/pangolin/utils/posix/shared_memory_buffer.h
-- Installing: /usr/local/include/pangolin/utils/posix/semaphore.h
-- Installing: /usr/local/include/pangolin/utils/registration.h
-- Installing: /usr/local/include/pangolin/utils/xml
-- Installing: /usr/local/include/pangolin/utils/xml/rapidxml_utils.hpp
-- Installing: /usr/local/include/pangolin/utils/xml/license.txt
-- Installing: /usr/local/include/pangolin/utils/xml/rapidxml_iterators.hpp
-- Installing: /usr/local/include/pangolin/utils/xml/rapidxml.hpp
-- Installing: /usr/local/include/pangolin/utils/xml/rapidxml_print.hpp
-- Installing: /usr/local/include/pangolin/utils/uri.h
-- Installing: /usr/local/include/pangolin/utils/compontent_cast.h
-- Installing: /usr/local/include/pangolin/plot
-- Installing: /usr/local/include/pangolin/plot/plotter.h
-- Installing: /usr/local/include/pangolin/plot/range.h
-- Installing: /usr/local/include/pangolin/plot/datalog.h
-- Installing: /usr/local/include/pangolin/display
-- Installing: /usr/local/include/pangolin/display/display.h
-- Installing: /usr/local/include/pangolin/display/device
-- Installing: /usr/local/include/pangolin/display/device/display_android.h
-- Installing: /usr/local/include/pangolin/display/device/X11Window.h
-- Installing: /usr/local/include/pangolin/display/device/PangolinNSApplication.h
-- Installing: /usr/local/include/pangolin/display/device/X11GlContext.h
-- Installing: /usr/local/include/pangolin/display/device/PangolinNSGLView.h
-- Installing: /usr/local/include/pangolin/display/device/OsxWindow.h
-- Installing: /usr/local/include/pangolin/display/device/WinWindow.h
-- Installing: /usr/local/include/pangolin/display/attach.h
-- Installing: /usr/local/include/pangolin/display/widgets
-- Installing: /usr/local/include/pangolin/display/widgets/widgets.h
-- Installing: /usr/local/include/pangolin/display/viewport.h
-- Installing: /usr/local/include/pangolin/display/user_app.h
-- Installing: /usr/local/include/pangolin/display/window.h
-- Installing: /usr/local/include/pangolin/display/opengl_render_state.h
-- Installing: /usr/local/include/pangolin/display/image_view.h
-- Installing: /usr/local/include/pangolin/display/display_internal.h
-- Installing: /usr/local/include/pangolin/display/view.h
-- Installing: /usr/local/include/pangolin/ios
-- Installing: /usr/local/include/pangolin/ios/PangolinUIView.h
-- Installing: /usr/local/include/pangolin/ios/PangolinAppDelegate.h
-- Installing: /usr/local/include/pangolin/factory
-- Installing: /usr/local/include/pangolin/factory/factory_registry.h
-- Installing: /usr/local/include/pangolin/python
-- Installing: /usr/local/include/pangolin/python/pyinterpreter.h
-- Installing: /usr/local/include/pangolin/python/pypangolin_init.h
-- Installing: /usr/local/include/pangolin/python/pyuniqueobj.h
-- Installing: /usr/local/include/pangolin/python/pyvar.h
-- Installing: /usr/local/include/pangolin/python/pypangoio.h
-- Installing: /usr/local/include/pangolin/geometry
-- Installing: /usr/local/include/pangolin/geometry/glgeometry.h
-- Installing: /usr/local/include/pangolin/geometry/geometry_obj.h
-- Installing: /usr/local/include/pangolin/geometry/geometry_ply.h
-- Installing: /usr/local/include/pangolin/geometry/geometry.h
-- Installing: /usr/local/include/pangolin/pangolin.h
-- Installing: /usr/local/lib/libpangolin.so
-- Installing: /usr/local/lib/cmake/Pangolin/PangolinConfig.cmake
-- Installing: /usr/local/lib/cmake/Pangolin/PangolinConfigVersion.cmake
-- Installing: /usr/local/lib/cmake/Pangolin/PangolinTargets.cmake
-- Installing: /usr/local/lib/cmake/Pangolin/PangolinTargets-release.cmake
-- Installing: /usr/local/bin/VideoViewer
-- Set runtime path of "/usr/local/bin/VideoViewer" to ""
-- Installing: /usr/local/bin/VideoConvert
-- Set runtime path of "/usr/local/bin/VideoConvert" to ""
-- Installing: /usr/local/bin/VideoJsonPrint
-- Set runtime path of "/usr/local/bin/VideoJsonPrint" to ""
-- Installing: /usr/local/bin/VideoJsonTransform
-- Set runtime path of "/usr/local/bin/VideoJsonTransform" to ""
-- Installing: /usr/local/bin/Plotter
-- Set runtime path of "/usr/local/bin/Plotter" to ""
ht_llibra@ht-llibra:~/VSlam/Pangolin-v0.6/build$

image-20250107203007538

在你所放置的Pangolin文件下

image-20250107203657314

cd build/examples/HelloPangolin
./HelloPangolin

运行此示例，效果显示是一个红绿蓝立方体，表示即为安装成功。

（这次在ubantu18.04安装出奇的顺利）

image-20250107203501549

3、Opencv安装

最新的ORB-SLAM3要求OpenCV 4.4及以上，但是最好不要安装OpenCV4以上，具体原因不太理解，最后选择安装3.4.3版本。首先，到官网下载源码：https://opencv.org/releases/page/5/，实验可行。
参考：https://blog.csdn.net/shadowmimii/article/details/129472719

unzip opencv-3.4.3.zip

image-20250107204807283

更新一下索引 sudo apt-get update，

使用一条指令安装所有依赖，

sudo apt-get install build-essential libgtk2.0-dev libavcodec-dev libavformat-dev libjpeg-dev libtiff5-dev libswscale-dev libjasper-dev

可能会出现以下报错信息，

ht_llibra@ht-llibra:~$ sudo apt-get install build-essential libgtk2.0-dev libavcodec-dev libavformat-dev libjpeg-dev libtiff5-dev libswscale-dev libjasper-dev
正在读取软件包列表... 完成
正在分析软件包的依赖关系树       
正在读取状态信息... 完成       
E: 无法定位软件包 libjasper-dev
ht_llibra@ht-llibra:~$

解决方法：执行以下命令之后再次安装

sudo add-apt-repository "deb http://security.ubuntu.com/ubuntu xenial-security main"
sudo apt update
sudo apt install libjasper1 libjasper-dev

sudo apt-get install build-essential libgtk2.0-dev libavcodec-dev libavformat-dev libjpeg.dev libtiff4.dev libswscale-dev libjasper-dev

不出意外，安装好依赖，开始编译OpenCV,

在opencv-3.4.3目录下，

mkdir build
cd build
cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local ..


--   Python (for build):            /usr/bin/python2.7
-- 
--   Java:                          
--     ant:                         NO
--     JNI:                         NO
--     Java wrappers:               NO
--     Java tests:                  NO
-- 
--   Matlab:                        NO
-- 
--   Install to:                    /usr/local
-- -----------------------------------------------------------------
-- 
-- Configuring done
-- Generating done
-- Build files have been written to: /home/ht_llibra/VSlam/opencv-3.4.3/build
ht_llibra@ht-llibra:~/VSlam/opencv-3.4.3/build$

不得不感叹一句，好顺利啊！还得是ubantu18.04。 记：需要科学上网，否则可能中间下载失败。

make -j4

j4表示4核，可以根据虚拟机分配的核数进行调整，也可以不用，直接make，速度稍微慢点。

时间肯有点长，编译完成之后需要再安装一下，

sudo make install

– Installing: /usr/local/bin/opencv_version
– Set runtime path of “/usr/local/bin/opencv_version” to “/usr/local/lib”

4、DBoW2 and g2o

DBoW2主要用于回环检测，g2o（General Graph Optimization）主要用于图优化。
ORB-SLAM3的源码包自带DBoW2 and g2o，编译时会自动安装，不用管。

外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传

5、编译ORB_slam3

参考：https://github.com/UZ-SLAMLab/ORB_SLAM3

源码下载：https://github.com/UZ-SLAMLab/ORB_SLAM3

git clone https://github.com/UZ-SLAMLab/ORB_SLAM3.git

打开ORB_SLAM3中build.sh可以看到5个需要编译的第三方库，

image-20250107212922797

直接执行以下命令，可一次性编译这5个文档，但如果报错不易查找问题,

chmod +x build.sh
./build.sh

因此，建议单独进行编译。

（1）编译Thirdparty/DBoW2

在ORB_SLAM3文件夹下打开终端

cd Thirdparty/DBoW2
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make -j4

（2）编译Thirdparty/g2o

在/ORB_SLAM3下开终端

cd Thirdparty/g2o
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make -j4

（3）编译Thirdparty/Sophus

在/ORB_SLAM3下开终端

cd Thirdparty/ Sophus
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make -j4

（4）解压/ORB_SLAM3/Vocabulary/ORBvoc.txt.tar.gz

在/ORB_SLAM3下开终端

cd Vocabulary
tar -xf ORBvoc.txt.tar.gz
cd ..

（5）编译ORB_SLAM3

mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make -j4

ht_llibra@ht-llibra:~/VSlam/ORB_SLAM3/build$ cmake .. -DCMAKE_BUILD_TYPE=Release-- The C compiler identification is GNU 7.5.0
-- The CXX compiler identification is GNU 7.5.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
Build type: Release
-- Performing Test COMPILER_SUPPORTS_CXX11
-- Performing Test COMPILER_SUPPORTS_CXX11 - Success
-- Performing Test COMPILER_SUPPORTS_CXX0X
-- Performing Test COMPILER_SUPPORTS_CXX0X - Success
-- Using flag -std=c++11.
CMake Warning at CMakeLists.txt:33 (find_package):
  Could not find a configuration file for package "OpenCV" that is compatible
  with requested version "4.4".

  The following configuration files were considered but not accepted:

    /usr/local/share/OpenCV/OpenCVConfig.cmake, version: 3.4.3
    /usr/share/OpenCV/OpenCVConfig.cmake, version: 3.2.0



CMake Error at CMakeLists.txt:35 (message):
  OpenCV > 4.4 not found.


-- Configuring incomplete, errors occurred!
See also "/home/ht_llibra/VSlam/ORB_SLAM3/build/CMakeFiles/CMakeOutput.log".
ht_llibra@ht-llibra:~/VSlam/ORB_SLAM3/build$ make -j4
make: *** 没有指明目标并且找不到 makefile。 停止。
ht_llibra@ht-llibra:~/VSlam/ORB_SLAM3/build$

报错原因：此处我安装的是opencv3.4.2
解决办法：将/ORB_SLAM3/CMakeLists.txt中第33行find_package(OpenCV 4.4)此处的4.4改为3.4.2即可，

即 find_package(OpenCV 3.4.2)，然后再次编译。

又编译报错c++: internal compiler error：可能是内存不足或 CPU 资源过度使用

扩了容没有挂载成功，容量没变化

VMWare扩容

显示应用程序 --> 磁盘 --> 设置 调整大小

image-20250107221726255

还是make -j4报错

试试减少并行编译任务数，降低 -j 参数的值，

make -j2

还是报错。。。

检测系统资源：

内存使用情况 free -h
CPU 使用情况 top

系统资源也没问题。

make

[ 71%] Linking CXX executable …/Examples/Stereo-Inertial/stereo_inertial_tum_vi
[ 71%] Built target stereo_inertial_tum_vi
[100%] Built target g2o
ht_llibra@ht-llibra:~/VSlam/ORB_SLAM3/build$

速度很慢，好在编译成功。

6、数据集下载及运行，评估

下载

https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets#downloads

image-20250107225514634

/ORB_SLAM3下新建Datasets,

将下载好的文件MH_01_easy放入Datasets，并改名为MH01，ROS bag也下载一个。

外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传

运行

参考另一个ORB_SLAM项目：https://github.com/j4mf/ORB_SLAM3 中的ORB_SLAM3/euroc_examples.sh ，

或者 https://github.com/electech6/ORB_SLAM3_detailed_comments 这个，一样的 ，下载后放入目录，

image-20250107233528324

#!/bin/bash
pathDatasetEuroc='/Datasets/EuRoC' #Example, it is necesary to change it by the dataset path

#------------------------------------
# Monocular Examples
echo "Launching MH01 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Monocular/EuRoC_TimeStamps/MH01.txt dataset-MH01_mono

echo "Launching MH02 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH02 ./Examples/Monocular/EuRoC_TimeStamps/MH02.txt dataset-MH02_mono

echo "Launching MH03 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH03 ./Examples/Monocular/EuRoC_TimeStamps/MH03.txt dataset-MH03_mono

echo "Launching MH04 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH04 ./Examples/Monocular/EuRoC_TimeStamps/MH04.txt dataset-MH04_mono

echo "Launching MH05 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH05 ./Examples/Monocular/EuRoC_TimeStamps/MH05.txt dataset-MH05_mono

echo "Launching V101 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Monocular/EuRoC_TimeStamps/V101.txt dataset-V101_mono

echo "Launching V102 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V102 ./Examples/Monocular/EuRoC_TimeStamps/V102.txt dataset-V102_mono

echo "Launching V103 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V103 ./Examples/Monocular/EuRoC_TimeStamps/V103.txt dataset-V103_mono

echo "Launching V201 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Monocular/EuRoC_TimeStamps/V201.txt dataset-V201_mono

echo "Launching V202 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V202 ./Examples/Monocular/EuRoC_TimeStamps/V202.txt dataset-V202_mono

echo "Launching V203 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V203 ./Examples/Monocular/EuRoC_TimeStamps/V203.txt dataset-V203_mono

# MultiSession Monocular Examples
echo "Launching Machine Hall with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Monocular/EuRoC_TimeStamps/MH01.txt "$pathDatasetEuroc"/MH02 ./Examples/Monocular/EuRoC_TimeStamps/MH02.txt "$pathDatasetEuroc"/MH03 ./Examples/Monocular/EuRoC_TimeStamps/MH03.txt "$pathDatasetEuroc"/MH04 ./Examples/Monocular/EuRoC_TimeStamps/MH04.txt "$pathDatasetEuroc"/MH05 ./Examples/Monocular/EuRoC_TimeStamps/MH05.txt dataset-MH01_to_MH05_mono

echo "Launching Vicon Room 1 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Monocular/EuRoC_TimeStamps/V101.txt "$pathDatasetEuroc"/V102 ./Examples/Monocular/EuRoC_TimeStamps/V102.txt "$pathDatasetEuroc"/V103 ./Examples/Monocular/EuRoC_TimeStamps/V103.txt dataset-V101_to_V103_mono

echo "Launching Vicon Room 2 with Monocular sensor"
./Examples/Monocular/mono_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Monocular/EuRoC_TimeStamps/V201.txt "$pathDatasetEuroc"/V202 ./Examples/Monocular/EuRoC_TimeStamps/V202.txt "$pathDatasetEuroc"/V203 ./Examples/Monocular/EuRoC_TimeStamps/V203.txt dataset-V201_to_V203_mono

#------------------------------------
# Stereo Examples
echo "Launching MH01 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo/EuRoC_TimeStamps/MH01.txt dataset-MH01_stereo

echo "Launching MH02 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH02 ./Examples/Stereo/EuRoC_TimeStamps/MH02.txt dataset-MH02_stereo

echo "Launching MH03 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH03 ./Examples/Stereo/EuRoC_TimeStamps/MH03.txt dataset-MH03_stereo

echo "Launching MH04 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH04 ./Examples/Stereo/EuRoC_TimeStamps/MH04.txt dataset-MH04_stereo

echo "Launching MH05 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH05 ./Examples/Stereo/EuRoC_TimeStamps/MH05.txt dataset-MH05_stereo

echo "Launching V101 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Stereo/EuRoC_TimeStamps/V101.txt dataset-V101_stereo

echo "Launching V102 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V102 ./Examples/Stereo/EuRoC_TimeStamps/V102.txt dataset-V102_stereo

echo "Launching V103 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V103 ./Examples/Stereo/EuRoC_TimeStamps/V103.txt dataset-V103_stereo

echo "Launching V201 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Stereo/EuRoC_TimeStamps/V201.txt dataset-V201_stereo

echo "Launching V202 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V202 ./Examples/Stereo/EuRoC_TimeStamps/V202.txt dataset-V202_stereo

echo "Launching V203 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V203 ./Examples/Stereo/EuRoC_TimeStamps/V203.txt dataset-V203_stereo

# MultiSession Stereo Examples
echo "Launching Machine Hall with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo/EuRoC_TimeStamps/MH01.txt "$pathDatasetEuroc"/MH02 ./Examples/Stereo/EuRoC_TimeStamps/MH02.txt "$pathDatasetEuroc"/MH03 ./Examples/Stereo/EuRoC_TimeStamps/MH03.txt "$pathDatasetEuroc"/MH04 ./Examples/Stereo/EuRoC_TimeStamps/MH04.txt "$pathDatasetEuroc"/MH05 ./Examples/Stereo/EuRoC_TimeStamps/MH05.txt dataset-MH01_to_MH05_stereo

echo "Launching Vicon Room 1 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Stereo/EuRoC_TimeStamps/V101.txt "$pathDatasetEuroc"/V102 ./Examples/Stereo/EuRoC_TimeStamps/V102.txt "$pathDatasetEuroc"/V103 ./Examples/Stereo/EuRoC_TimeStamps/V103.txt dataset-V101_to_V103_stereo

echo "Launching Vicon Room 2 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Stereo/EuRoC_TimeStamps/V201.txt "$pathDatasetEuroc"/V202 ./Examples/Stereo/EuRoC_TimeStamps/V202.txt "$pathDatasetEuroc"/V203 ./Examples/Stereo/EuRoC_TimeStamps/V203.txt dataset-V201_to_V203_stereo

#------------------------------------
# Monocular-Inertial Examples
echo "Launching MH01 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH01.txt dataset-MH01_monoi

echo "Launching MH02 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH02 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH02.txt dataset-MH02_monoi

echo "Launching MH03 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH03 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH03.txt dataset-MH03_monoi

echo "Launching MH04 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH04 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH04.txt dataset-MH04_monoi

echo "Launching MH05 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH05 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH05.txt dataset-MH05_monoi

echo "Launching V101 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V101.txt dataset-V101_monoi

echo "Launching V102 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V102 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V102.txt dataset-V102_monoi

echo "Launching V103 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V103 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V103.txt dataset-V103_monoi

echo "Launching V201 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V201.txt dataset-V201_monoi

echo "Launching V202 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V202 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V202.txt dataset-V202_monoi

echo "Launching V203 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V203 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V203.txt dataset-V203_monoi

# MultiSession Monocular Examples
echo "Launching Machine Hall with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH01.txt "$pathDatasetEuroc"/MH02 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH02.txt "$pathDatasetEuroc"/MH03 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH03.txt "$pathDatasetEuroc"/MH04 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH04.txt "$pathDatasetEuroc"/MH05 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/MH05.txt dataset-MH01_to_MH05_monoi

echo "Launching Vicon Room 1 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V101.txt "$pathDatasetEuroc"/V102 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V102.txt "$pathDatasetEuroc"/V103 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V103.txt dataset-V101_to_V103_monoi

echo "Launching Vicon Room 2 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V201.txt "$pathDatasetEuroc"/V202 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V202.txt "$pathDatasetEuroc"/V203 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V203.txt dataset-V201_to_V203_monoi

#------------------------------------
# Stereo-Inertial Examples
echo "Launching MH01 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH01.txt dataset-MH01_stereoi

echo "Launching MH02 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH02 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH02.txt dataset-MH02_stereoi

echo "Launching MH03 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH03 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH03.txt dataset-MH03_stereoi

echo "Launching MH04 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH04 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH04.txt dataset-MH04_stereoi

echo "Launching MH05 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH05 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH05.txt dataset-MH05_stereoi

echo "Launching V101 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V101.txt dataset-V101_stereoi

echo "Launching V102 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V102 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V102.txt dataset-V102_stereoi

echo "Launching V103 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V103 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V103.txt dataset-V103_stereoi

echo "Launching V201 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V201.txt dataset-V201_stereoi

echo "Launching V202 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V202 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V202.txt dataset-V202_stereoi

echo "Launching V203 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V203 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V203.txt dataset-V203_stereoi

# MultiSession Stereo-Inertial Examples
echo "Launching Machine Hall with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH01.txt "$pathDatasetEuroc"/MH02 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH02.txt "$pathDatasetEuroc"/MH03 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH03.txt "$pathDatasetEuroc"/MH04 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH04.txt "$pathDatasetEuroc"/MH05 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/MH05.txt dataset-MH01_to_MH05_stereoi

echo "Launching Vicon Room 1 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V101 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V101.txt "$pathDatasetEuroc"/V102 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V102.txt "$pathDatasetEuroc"/V103 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V103.txt dataset-V101_to_V103_stereoi

echo "Launching Vicon Room 2 with Stereo-Inertial sensor"
./Examples/Stereo-Inertial/stereo_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V201.txt "$pathDatasetEuroc"/V202 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V202.txt "$pathDatasetEuroc"/V203 ./Examples/Stereo-Inertial/EuRoC_TimeStamps/V203.txt dataset-V201_to_V203_stereoi

找到MH01对应sensor的命令运行即可，如：

 ./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml /home/ht_llibra/VSlam/ORB_SLAM3/Datasets/MH01 ./Examples/Stereo/EuRoC_TimeStamps/MH01.txt dataset-MH01_stereo

注意路径更改。

image-20250107234346387

image-20250107234249940

评估

评估脚本euroc_eval_examples.sh，

#!/bin/bash
pathDatasetEuroc='/Datasets/EuRoC' #Example, it is necesary to change it by the dataset path

# Single Session Example (Pure visual)
echo "Launching MH01 with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo/EuRoC_TimeStamps/MH01.txt dataset-MH01_stereo
echo "------------------------------------"
echo "Evaluation of MH01 trajectory with Stereo sensor"
python evaluation/evaluate_ate_scale.py evaluation/Ground_truth/EuRoC_left_cam/MH01_GT.txt f_dataset-MH01_stereo.txt --plot MH01_stereo.pdf


# MultiSession Example (Pure visual)
echo "Launching Machine Hall with Stereo sensor"
./Examples/Stereo/stereo_euroc ./Vocabulary/ORBvoc.txt ./Examples/Stereo/EuRoC.yaml "$pathDatasetEuroc"/MH01 ./Examples/Stereo/EuRoC_TimeStamps/MH01.txt "$pathDatasetEuroc"/MH02 ./Examples/Stereo/EuRoC_TimeStamps/MH02.txt "$pathDatasetEuroc"/MH03 ./Examples/Stereo/EuRoC_TimeStamps/MH03.txt "$pathDatasetEuroc"/MH04 ./Examples/Stereo/EuRoC_TimeStamps/MH04.txt "$pathDatasetEuroc"/MH05 ./Examples/Stereo/EuRoC_TimeStamps/MH05.txt dataset-MH01_to_MH05_stereo
echo "------------------------------------"
echo "Evaluation of MAchine Hall trajectory with Stereo sensor"
python evaluation/evaluate_ate_scale.py evaluation/Ground_truth/EuRoC_left_cam/MH_GT.txt f_dataset-MH01_to_MH05_stereo.txt --plot MH01_to_MH05_stereo.pdf


# Single Session Example (Visual-Inertial)
echo "Launching V102 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V102 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V102.txt dataset-V102_monoi
echo "------------------------------------"
echo "Evaluation of V102 trajectory with Monocular-Inertial sensor"
python evaluation/evaluate_ate_scale.py "$pathDatasetEuroc"/V102/mav0/state_groundtruth_estimate0/data.csv f_dataset-V102_monoi.txt --plot V102_monoi.pdf


# MultiSession Monocular Examples

echo "Launching Vicon Room 2 with Monocular-Inertial sensor"
./Examples/Monocular-Inertial/mono_inertial_euroc ./Vocabulary/ORBvoc.txt ./Examples/Monocular-Inertial/EuRoC.yaml "$pathDatasetEuroc"/V201 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V201.txt "$pathDatasetEuroc"/V202 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V202.txt "$pathDatasetEuroc"/V203 ./Examples/Monocular-Inertial/EuRoC_TimeStamps/V203.txt dataset-V201_to_V203_monoi
echo "------------------------------------"
echo "Evaluation of Vicon Room 2 trajectory with Stereo sensor"
python evaluation/evaluate_ate_scale.py evaluation/Ground_truth/EuRoC_imu/V2_GT.txt f_dataset-V201_to_V203_monoi.txt --plot V201_to_V203_monoi.pdf

执行脚本，

./euroc_examples.sh

部分终端记录：

Loading ORB Vocabulary. This could take a while...
Vocabulary loaded!

Initialization of Atlas from scratch 
Creation of new map with id: 0
Creation of new map with last KF id: 0
Seq. Name: 
There are 1 cameras in the atlas
Camera 0 is pinhole

Failed to load image at: /Datasets/EuRoC/V101/mav0/cam0/data/1403715273262142976.png
Launching Vicon Room 2 with Stereo sensor
num_seq = 3
file name: dataset-V201_to_V203_stereo
Loading images for sequence 0...LOADED!
Loading images for sequence 1...LOADED!
Loading images for sequence 2...LOADED!

-------

ORB-SLAM3 Copyright (C) 2017-2020 Carlos Campos, Richard Elvira, Juan J. Gómez, José M.M. Montiel and Juan D. Tardós, University of Zaragoza.
ORB-SLAM2 Copyright (C) 2014-2016 Raúl Mur-Artal, José M.M. Montiel and Juan D. Tardós, University of Zaragoza.
This program comes with ABSOLUTELY NO WARRANTY;
This is free software, and you are welcome to redistribute it
under certain conditions. See LICENSE.txt.

Input sensor was set to: Stereo
Loading settings from ./Examples/Stereo/EuRoC.yaml
Camera1.k3 optional parameter does not exist...
	-Loaded camera 1
Camera2.k3 optional parameter does not exist...
	-Loaded camera 2
Camera.newHeight optional parameter does not exist...
Camera.newWidth optional parameter does not exist...
	-Loaded image info
	-Loaded ORB settings
	-Loaded viewer settings
System.LoadAtlasFromFile optional parameter does not exist...
System.SaveAtlasToFile optional parameter does not exist...
	-Loaded Atlas settings
System.thFarPoints optional parameter does not exist...
	-Loaded misc parameters
	-Computed rectification maps
----------------------------------
SLAM settings: 
	-Camera 1 parameters (Pinhole): [ 458.65399169921875 457.29598999023438 367.21499633789062 248.375 ]
	-Camera 1 distortion parameters: [  -0.28340810537338257 0.073959067463874817 0.00019359000725671649 1.7618711353861727e-05 ]
	-Camera 2 parameters (Pinhole: [ 457.58700561523438 456.13400268554688 379.99899291992188 255.23800659179688 ]
	-Camera 1 distortion parameters: [  -0.2836836576461792 0.074512839317321777 -0.00010473000293131918 -3.555907096597366e-05 ]
	-Original image size: [ 752 , 480 ]
	-Current image size: [ 752 , 480 ]
	-Camera 1 parameters after rectification: [  456.71499633789062 456.71499633789062 364.28717041015625 257.04702758789062 ]
	-Sequence FPS: 20
	-Stereo baseline: 0.11007784307003021
	-Stereo depth threshold : 60
	-Features per image: 1200
	-ORB scale factor: 1.2000000476837158
	-ORB number of scales: 8
	-Initial FAST threshold: 20
	-Min FAST threshold: 7

7、运行相机（用的是 Intel D455）

sudo apt-get install ros-melodic-realsense2-camera

ht_llibra@ht-llibra:~$ rospack find realsense2_camera
/opt/ros/melodic/share/realsense2_camera

这里咱们直接用apt安装，之前在ubantu20.04和ubantu22.04用过源码编译安装。

https://github.com/IntelRealSense/realsense-ros/tree/ros1-legacy

https://github.com/IntelRealSense/realsense-ros/tree/ros2-master

安装好后，连接相机，测试摄像头能否正确打开，

roslaunch realsense2_camera rs_rgbd.launch
roslaunch realsense2_camera rs_camera.launch
roslaunch realsense2_camera rs_camera.launch align_depth:=true
roslaunch realsense2_camera rs_camera.launch filters:=pointcloud

报错！解决：相机需要使用USB3，虚拟机默认配置的USB2

image-20250108004325561

不知道为啥就是用不了，咱还是用源码编译吧！

参考：https://blog.csdn.net/u014374826/article/details/132000684

（1）安装librealsense SDK2.0

首先·注册秘钥：

sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-key F6E65AC044F831AC80A06380C8B3A55A6F3EFCDE

添加服务器地址：

sudo add-apt-repository "deb https://librealsense.intel.com/Debian/apt-repo $(lsb_release -cs) main" -u

安装必要库：

sudo apt-get install librealsense2-dkms
sudo apt-get install librealsense2-utils

打开显示界面验证：

realsense-viewer

image-20250108143037193

这里没连接相机，连接后会有画面，注意这里虚拟机需要改USB3.0以上才能连接成功，测试可用。

再次验证：

ht_llibra@ht-llibra:~$ modinfo uvcvideo | grep "version:"
version:        1.1.2.realsense-1.3.18
srcversion:     51A4A0210E91BE50A7BA2BB
ht_llibra@ht-llibra:~$

检测摄像头是否被正确识别：

rs-enumerate-devices

（2）安装realsense-ros

先将前面apt安装的卸载，避免后续冲突，

sudo apt-get remove ros-melodic-realsense2-camera

如需彻底移除配置文件和其他相关数据，可以使用 purge 命令,

sudo apt-get purge ros-melodic-realsense2-camera

这里先使用

ht_llibra@ht-llibra:~$ sudo apt-get remove ros-melodic-realsense2-camera
正在读取软件包列表... 完成
正在分析软件包的依赖关系树       
正在读取状态信息... 完成       
下列软件包是自动安装的并且现在不需要了：
  ros-melodic-ddynamic-reconfigure ros-melodic-librealsense2
使用'sudo apt autoremove'来卸载它(它们)。
下列软件包将被【卸载】：
  ros-melodic-realsense2-camera
升级了 0 个软件包，新安装了 0 个软件包，要卸载 1 个软件包，有 56 个软件包未被升级。
解压缩后将会空出 1,412 kB 的空间。
您希望继续执行吗？ [Y/n] Y
(正在读取数据库 ... 系统当前共安装有 264442 个文件和目录。)
正在卸载 ros-melodic-realsense2-camera (2.3.2-1bionic.20221025.193216) ...
ht_llibra@ht-llibra:~$ 

ht_llibra@ht-llibra:~$ rospack find realsense2-camera
[rospack] Error: package 'realsense2-camera' not found
ht_llibra@ht-llibra:~$

image-20250108144818407

git clone https://github.com/IntelRealSense/realsense-ros.git  现在默认ros2分支，建议手动下载

初始化工作空间：

cd realsense_ws/src
catkin_init_workspace
cd ..
catkin_make clean

记录：

ht_llibra@ht-llibra:~/VSlam$ cd realsense_ws/src
ht_llibra@ht-llibra:~/VSlam/realsense_ws/src$ catkin_init_workspace
Creating symlink "/home/ht_llibra/VSlam/realsense_ws/src/CMakeLists.txt" pointing to "/opt/ros/melodic/share/catkin/cmake/toplevel.cmake"
ht_llibra@ht-llibra:~/VSlam/realsense_ws/src$ cd ..
ht_llibra@ht-llibra:~/VSlam/realsense_ws$ catkin_make clean

ht_llibra@ht-llibra:~/VSlam/realsense_ws$ catkin_make clean
Base path: /home/ht_llibra/VSlam/realsense_ws
Source space: /home/ht_llibra/VSlam/realsense_ws/src
Build space: /home/ht_llibra/VSlam/realsense_ws/build
Devel space: /home/ht_llibra/VSlam/realsense_ws/devel
Install space: /home/ht_llibra/VSlam/realsense_ws/install
####
#### Running command: "cmake /home/ht_llibra/VSlam/realsense_ws/src -DCATKIN_DEVEL_PREFIX=/home/ht_llibra/VSlam/realsense_ws/devel -DCMAKE_INSTALL_PREFIX=/home/ht_llibra/VSlam/realsense_ws/install -G Unix Makefiles" in "/home/ht_llibra/VSlam/realsense_ws/build"
####
-- The C compiler identification is GNU 7.5.0
-- The CXX compiler identification is GNU 7.5.0
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Using CATKIN_DEVEL_PREFIX: /home/ht_llibra/VSlam/realsense_ws/devel
-- Using CMAKE_PREFIX_PATH: /home/ht_llibra/carto_ws/cartographer_detailed_comments_ws/install_isolated;/opt/ros/melodic
-- This workspace overlays: /home/ht_llibra/carto_ws/cartographer_detailed_comments_ws/install_isolated;/opt/ros/melodic
-- Found PythonInterp: /usr/bin/python2 (found suitable version "2.7.17", minimum required is "2") 
-- Using PYTHON_EXECUTABLE: /usr/bin/python2
-- Using Debian Python package layout
-- Using empy: /usr/bin/empy
-- Using CATKIN_ENABLE_TESTING: ON
-- Call enable_testing()
-- Using CATKIN_TEST_RESULTS_DIR: /home/ht_llibra/VSlam/realsense_ws/build/test_results
-- Found gtest sources under '/usr/src/googletest': gtests will be built
-- Found gmock sources under '/usr/src/googletest': gmock will be built
-- Found PythonInterp: /usr/bin/python2 (found version "2.7.17") 
-- Looking for pthread.h
-- Looking for pthread.h - found
-- Looking for pthread_create
-- Looking for pthread_create - not found
-- Looking for pthread_create in pthreads
-- Looking for pthread_create in pthreads - not found
-- Looking for pthread_create in pthread
-- Looking for pthread_create in pthread - found
-- Found Threads: TRUE  
-- Using Python nosetests: /usr/bin/nosetests-2.7
-- catkin 0.7.29
-- BUILD_SHARED_LIBS is on
-- BUILD_SHARED_LIBS is on
-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-- ~~  traversing 2 packages in topological order:
-- ~~  - realsense2_description
-- ~~  - realsense2_camera
-- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-- +++ processing catkin package: 'realsense2_description'
-- ==> add_subdirectory(realsense-ros/realsense2_description)
-- +++ processing catkin package: 'realsense2_camera'
-- ==> add_subdirectory(realsense-ros/realsense2_camera)
-- Using these message generators: gencpp;geneus;genlisp;gennodejs;genpy
-- Create Release Build.
-- realsense2_camera: 3 messages, 1 services
-- Configuring done
-- Generating done
-- Build files have been written to: /home/ht_llibra/VSlam/realsense_ws/build
####
#### Running command: "make clean -j4 -l4" in "/home/ht_llibra/VSlam/realsense_ws/build"
####
ht_llibra@ht-llibra:~/VSlam/realsense_ws$

没有报错，之前在ubantu20.04和22.04都出现过报错，肯是前面apt安装的一些依赖没卸载，此处用到了，反正很顺利。

开始安装源码: 回到工作空间

catkin_make -DCATKIN_ENABLE_TESTING=False -DCMAKE_BUILD_TYPE=Release
catkin_make install

添加到系统环境变量：

image-20250108222057563

echo "source /home/ht_llibra/VSlam/realsense_ws/devel/setup.bash" >> ~/.bashrc
source ~/.bashrc

验证是否安装成功：

roslaunch realsense2_camera rs_camera.launch

报错没显示画面,

 08/01 20:24:50,782 WARNING [139676547536640] (messenger-libusb.cpp:42) control_transfer returned error, index: 768, error: Resource temporarily unavailable, number: 11
 08/01 20:24:51,348 WARNING [139676547536640] (messenger-libusb.cpp:42) control_transfer returned error, index: 768, error: Resource temporarily unavailable, number: 11
 08/01 20:25:02,952 WARNING [139676547536640] (messenger-libusb.cpp:42) control_transfer returned error, index: 768, error: Resource temporarily unavailable, number: 11

这个问题可能是librealsense版本和realsense-ros版本所需版本不一致！

参考：在配置Intel realsense (D435i)时遇到的问题（转载）_ubuntu20.04 d435i没有imu话题-CSDN博客

image-20250108203757461

查看README.md文件，LibRealSense2 supported version: v2.50.0 ，

然而apt默认安装的librealsense版本为2.55.1（realsense-viewer打开可见），这里可能需要卸载重装，

https://github.com/IntelRealSense/librealsense/releases/tag/v2.50.0

Release Intel® RealSense™ SDK 2.0 (v2.48.0) · IntelRealSense/librealsense · GitHub

Ubuntu18.04安装librealsense（D455相机）_虚拟机ubuntu下载librealsense-CSDN博客

然后我又发现可能是我想错了，打不开是因为环境变量没刷新（但是我明明加入了系统环境变量），修改为相对路径，（后续又发现是apt安装的realsense-ros没删干净，用命令安装realsense-ros也会同时安装librealsense2.50.0，所以可以用，删了之后出现版本不一致问题，详情见手持建图笔记）

source ~/VSlam/realsense_ws/devel/setup.bash

前面加的这个source /home/ht_llibra/VSlam/realsense_ws/devel/setup.bash不知道为啥刷新不了，明明是pwd打印的绝对路径。

image-20250108220622884

回顾以前安装的：

（1）Ubantu22.04虚拟机，ros2版本的，支持最新版librealsense，发现也是2.55.1 （具体安装过程之前没有记录下来），

image-20250108211701743

ros2 launch realsense2_camera rs_launch.py camera_namespace:=robot1 camera_name:=D455_1

rviz2   注意：sudo apt install ros-humble-rviz2

使用正常。

外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传

（2）Ubantu20.04双系统，ros1版本的，

和上述ubantu18.04虚拟机安装的一样，librealsense版本也为2.55.1，但是功能正常。

roslaunch realsense2_camera rs_rgbd.launch

外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传

解决方法：在终端中输入以下指令，安装rgbd_launch的功能包依赖

sudo apt-get install ros-melodic-rgbd-launch

启动相机节点并使用 pointcloud 选项发布点云：

roslaunch realsense2_camera rs_camera.launch filters:=pointcloud

启动相机节点并发布对齐的深度流：

roslaunch realsense2_camera rs_camera.launch align_depth:=true

使用 rqt_reconfigure 调整相机参数:

rosrun rqt_reconfigure rqt_reconfigure

7、ROS部分编译

原作者将ROS部分放到了/ORB_SLAM3/Examples_old位置，后续编译是进入old中。

在~/.bashrc中添加环境变量，将以下复制粘贴到最后一行：

export ROS_PACKAGE_PATH=${ROS_PACKAGE_PATH}:/home/ht_llibra/VSlam/ORB_SLAM3/Examples_old/ROS

开始编译：

cd Examples_old/ROS/ORB_SLAM3
mkdir build
cd build
cmake .. -DROS_BUILD_TYPE=Release
make -j4

或者直接运行脚本，但是这个路径不对，

image-20250108104125849

chmod +x build_ros.sh
./build_ros.sh

记：不小心使用错命令./build.sh又编译了一遍orbslam3，编译报错，但是还能用，所以应该没有覆盖，不知道有没有影响，后续有待观测。

因此还是单独编译，最后一步make -j4报错：

fatal error: sophus/se3.hpp: 没有那个文件或目录
#include <sophus/se3.hpp>，

image-20250108110648907

开始解决：

找到ORB_SLAM3/Examples_old/ROS/ORB_SLAM3下的CMakeLists.txt 修改它，include_directories中添加

${PROJECT_SOURCE_DIR}/../../../Thirdparty/Sophus

再次make -j4编译, 又有新的报错，

image-20250108111634977

（1）找到ORB_SLAM3/Examples_old/ROS/ORB_SLAM3/src/AR下的ros_mono_ar.cc文件：
将第151行的

cv::Mat Tcw = mpSLAM→TrackMonocular(cv_ptr→image,cv_ptr→header.stamp.toSec());

替换为

 cv::Mat Tcw;
 Sophus::SE3f  Tcw_SE3f = mpSLAM->TrackMonocular(cv_ptr->image,cv_ptr->header.stamp.toSec());
 Eigen::Matrix4f Tcw_Matrix = Tcw_SE3f.matrix();
 cv::eigen2cv(Tcw_Matrix, Tcw);

（2）找到ORB_SLAM3/Examples_old/ROS/ORB_SLAM3/src/AR下的ViewerAR.cc文件：
将第405行的

vPoints.push_back(pMP→GetWorldPos());

替换为

cv::Mat WorldPos;
cv::eigen2cv(pMP->GetWorldPos(), WorldPos);
vPoints.push_back(WorldPos);

将532行的

cv::Mat Xw = pMP→GetWorldPos();

替换为

cv::Mat Xw;
cv::eigen2cv(pMP->GetWorldPos(), Xw);

再次编译，还是报错

image-20250108112921473

加头文件：在/ORB_SLAM3/Examples_old/ROS/ORB_SLAM3/src/AR/ViewerAR.h中加入如下

#include <Eigen/Dense>
#include <opencv2/core/eigen.hpp>
#include <opencv2/opencv.hpp>

编译成功！！！

运行

roscore

在ORB_SLAM3下打开终端（每开一次都要source）

source /home/ht_llibra/VSlam/ORB_SLAM3/Examples_old/ROS/ORB_SLAM3/build/devel/setup.bash

rosrun ORB_SLAM3 Stereo Vocabulary/ORBvoc.txt Examples_old/Stereo/EuRoC.yaml /home/ht_llibra/VSlam/ORB_SLAM3/Datasets/MH_01_easy.bag

· 第一个参数为系统的主程序执行文件 Stereo

· 第二个参数为特征词典 Vocabulary/ORBvoc.txt

· 第三个参数为相机参数 Examples_old/Stereo/EuRoC.yaml
注意：后续如果要使用自己的数据集，是需要修改这个文件的。

· 第四个参数为数据集路径 /home/nj/下载/MH_01_easy.bag)

数据包回放： 启动双目 --pause默认暂停播放，按下空格键开始播放

rosbag play --pause --clock MH_01_easy.bag \
  /cam0/image_raw:=/camera/left/image_raw \
  /cam1/image_raw:=/camera/right/image_raw \
  /imu0:=/imu

发现显示画面不同步，上面的命令是启动双目的配置，但文件中缺少左右摄像头标定的内参，所以无法跟踪，需要的话就标了放进去编译启动。

roscore

source /home/ht_llibra/VSlam/ORB_SLAM3/Examples_old/ROS/ORB_SLAM3/build/devel/setup.bash

rosrun ORB_SLAM3 Mono Vocabulary/ORBvoc.txt Examples_old/Monocular-Inertial/EuRoC.yaml

数据包回放：启动双目 --pause默认暂停播放，按下空格键开始播放

rosbag play --pause MH_01_easy.bag /cam0/image_raw:=/camera/image_raw /imu0:=/imu

注：后面对应自己的话题，一定要和节点要求的对应。

image-20250108132754363

image-20250108132912056

查看话题：

rosnode info / Mono_Inertial
rospack list | grep image_transport

image-20250108132523294

ht_llibra@ht-llibra:~$ rospack list | grep image_transport
compressed_depth_image_transport /opt/ros/melodic/share/compressed_depth_image_transport
compressed_image_transport /opt/ros/melodic/share/compressed_image_transport
image_transport /opt/ros/melodic/share/image_transport
theora_image_transport /opt/ros/melodic/share/theora_image_transport
ht_llibra@ht-llibra:~$

PS: 踩坑小记

使用ROS编译ORB-SLAM2时遇到的问题（已解决）-CSDN博客

由于ORBSLAM中设置了rosbuild_init()，而我们编译文档之前需要先关联工作区间。ORBSLAM这里提供的这个文档中，并没有我们熟悉的ROS里的package.xml和devel这些文件，直接编译会显示ORBSLAM路径冲突之类的问题

但是我这里加了环境变量路径还是不行，最后发现必须加在最后一排才生效

# export ROS_PACKAGE_PATH=${ROS_PACKAGE_PATH}:/home/ht_llibra/VSlam/ORB_SLAM3/Examples_old/ROS
export ROS_PACKAGE_PATH=${ROS_PACKAGE_PATH}:~/VSlam/ORB_SLAM3/Examples_old/ROS

image-20250116221819632

再次编译，编译成功！

安装大佬注释修正版

https://github.com/electech6/ORB_SLAM3_detailed_comments

https://blog.csdn.net/Zoey_chn/article/details/135139494

git clone https://github.com/electech6/ORB_SLAM3_detailed_comments.git

image-20250108134252988

我用的opencv3.4.3，需要对应修改一下，

编译：

chmod +x build.sh 
./build.sh

安装前面那个后电脑内存不够了，还没试验。

ubantu20.04版本

案例1：

https://github.com/lturing/ORB_SLAM3_ROS

# 安装ros
# http://wiki.ros.org/noetic/Installation/Ubuntu
sudo apt install ros-noetic-desktop-full

git clone https://github.com/lturing/ORB_SLAM3_ROS.git
cd ORB_SLAM3_ROS
chmod +x ./build.sh 
./build.sh 

# 数据集下载
# https://www.cvlibs.net/datasets/kitti/eval_odometry.php
# https://cvg.cit.tum.de/data/datasets/rgbd-dataset/download

# 生成associate.txt(示例)
python evaluation/associate.py /home/spurs/dataset/rgbd_dataset_freiburg2_large_with_loop/rgb.txt /home/spurs/dataset/rgbd_dataset_freiburg2_large_with_loop/depth.txt >> associate.txt

# 运行单目
# 适当修改 launch/orb_slam_mono.launch
chmod +x ./run_mono.sh
./run_mono.sh 

# 运行双目
# 适当修改 launch/orb_slam_stereo.launch
chmod +x ./run_stereo.sh 
./run_stereo.sh 

# 运行rgbd稠密建图
# 适当修改 launch/orb_slam_rgbd_mapping.launch
chmod +x ./run_rgbd_mapping.sh 
./run_rgbd_mapping.sh

案例2：可视化工具

https://github.com/sunshanlu/SLAM_VIEWER

案例3：Yolov5_ORB_SLAM3

https://github.com/YWL0720/YOLO_ORB_SLAM3

https://github.com/triple-Mu/YOLOv8-TensorRT

案例4：vins-mono

https://blog.csdn.net/zardforever123/article/details/129042974

项目地址：GitHub - HKUST-Aerial-Robotics/VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator

VINS-Mono 论文地址： https://ieeexplore.ieee.org/abstract/document/8593603

https://ieeexplore.ieee.org/document/8421746?arnumber=8421746&source=authoralert

icle/details/135139494

git clone https://github.com/electech6/ORB_SLAM3_detailed_comments.git

[外链图片转存中…(img-3DvWGQ2T-1746537461615)]

我用的opencv3.4.3，需要对应修改一下，

编译：

chmod +x build.sh 
./build.sh

安装前面那个后电脑内存不够了，还没试验。

ubantu20.04版本

案例1：

https://github.com/lturing/ORB_SLAM3_ROS

# 安装ros
# http://wiki.ros.org/noetic/Installation/Ubuntu
sudo apt install ros-noetic-desktop-full

git clone https://github.com/lturing/ORB_SLAM3_ROS.git
cd ORB_SLAM3_ROS
chmod +x ./build.sh 
./build.sh 

# 数据集下载
# https://www.cvlibs.net/datasets/kitti/eval_odometry.php
# https://cvg.cit.tum.de/data/datasets/rgbd-dataset/download

# 生成associate.txt(示例)
python evaluation/associate.py /home/spurs/dataset/rgbd_dataset_freiburg2_large_with_loop/rgb.txt /home/spurs/dataset/rgbd_dataset_freiburg2_large_with_loop/depth.txt >> associate.txt

# 运行单目
# 适当修改 launch/orb_slam_mono.launch
chmod +x ./run_mono.sh
./run_mono.sh 

# 运行双目
# 适当修改 launch/orb_slam_stereo.launch
chmod +x ./run_stereo.sh 
./run_stereo.sh 

# 运行rgbd稠密建图
# 适当修改 launch/orb_slam_rgbd_mapping.launch
chmod +x ./run_rgbd_mapping.sh 
./run_rgbd_mapping.sh

案例2：可视化工具

https://github.com/sunshanlu/SLAM_VIEWER

案例3：Yolov5_ORB_SLAM3

https://github.com/YWL0720/YOLO_ORB_SLAM3

https://github.com/triple-Mu/YOLOv8-TensorRT

案例4：vins-mono

https://blog.csdn.net/zardforever123/article/details/129042974

项目地址：GitHub - HKUST-Aerial-Robotics/VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator

VINS-Mono 论文地址： https://ieeexplore.ieee.org/abstract/document/8593603

https://ieeexplore.ieee.org/document/8421746?arnumber=8421746&source=authoralert

呼啦月马
关注
Ubuntu18.04配置ORB_SLAM2运行环境
mw的博客
09-26 1209
（1）下载opencv3.4.1版本,安装版本过高会导致编译报错，下载地址：https://github.com/opencv/opencv。（1）下载ORB_SLAM2,放置在catkin_ws/src目录下，为后面在ROS 环境下运行做好准备。（2）在~/ORB_SLAM2/CMakeLists.txt文件中，改成自己的opencv版本。（2）然后上github，将0.5版本的下载下来（过高导致后面编译报错）（4）再次运行./build_ros.sh。（3）安装opencv3。
【ORB_SLAM系列2】—— 如何在Ubuntu18.04中安装运行ROS版本的ORB_SLAM3（踩坑记录！！！）
2401_82458959的博客
05-13 2438
以上就是运行ROS版本的ORB_SLAM的全部详细教程，主要是介绍了关于在ubuntu18.04中运行ORB_SLAM3的所有依赖的库的安装过程，和所遇到的一些报错的解决方法，最后也详细介绍如何下载，播放rosbag的形式，运行ORB_SLAM进行建图，下期分享如何在ubuntu18.04系统中使用自己的单目摄像头运行ORB_SLAM3。有什么问题，可以评论区里面交流讨论。
ubuntu18.04配置ORB-SLAM3（包含ROS）完整版教程
热门推荐
hhz_999的博客
09-25 2万+
ORB_SLAM3安装教程ORB_SLAM3安装准备1、C++11 or C++0x Compiler2、Pangolin3、OpenCV安装4、Eigen安装5、boost安装6、DBoW2 and g2o正式编译ORB_SLAM3测试ROS安装 本人是武汉大学测绘国重实验室的一名slam小白。组内有新生讲课的习惯，目前选题为视觉slam介绍，程序方面就选择orbslam3的编译和运行，本次配置花费了将近3天时间，遇到很多的bug，幸运的是每个问题都找到了原因并且配置成功了。 ps：只能说时代变化太快了，
（记录）Ubuntu18.04安装ORB-SLAM3过程
qq_39247667的博客
05-12 3259
Ubuntu18.04安装ORB-SLAM3过程记录Pangolin安装opencv安装Eigen 3安装Boost安装ORB_SLAM3的编译和安装数据集测试 Pangolin安装 Pangolin主要用于可视化和用户界面。 官方教程在: https://github.com/stevenlovegrove/Pangolin. //一些必要依赖 sudo apt install libglew-dev sudo apt install cmake sudo apt install libpython2.7
Ubuntu18.04安装和配置ORB_SLAM2（非ROS、ROS环境）
m0_65009717的博客
04-01 1万+
Ubuntu18.04下安装和配置ORB_SLAM2
自己ubuntu18.04上部署ORBSLAM3的可行操作步骤整理
诗筱涵的博客
07-21 982
OTB_SLAM3文件夹下的CMakeLists.txt里面添加下面这句（我不确定这步有没有起作用，后来似乎主要是靠更换ceres版本解决的编译报的一大堆语法错误，但是至少更换ceres版本再加上这步是没有问题的，可以解决那一大堆报错，不加这步只改ceres版本不知道是否OK，只加这步但是不改ceres版本是解决不了那些一大堆报错的，万全起见，这步还是加上，至少没坏处）其实到这里可以下数据集跑，当然不基于ROS的，但是由于下数据集的网站网速太慢，所以我就没有在这里下数据集跑了。
ORB_SLAM3在ubuntu18.04下的环境搭建及测试
YOULANSHENGMENG的博客
05-12 2669
参考的博文链接： Ubuntu18.04配置ORB SLAM3_爱吃鱼的森先生的博客-CSDN博客
ubuntu18.04-ORB-slam2安装与运行
honger17的博客
07-08 3442
ROS ORB-SLAM2安装教程
jetson nano ubuntu18.04 编译安装ORB_SLAM3
Isaachim的博客
09-11 1907
该文章与我另一篇文章类似，只不过该篇是在jetson nano上编译的方法， 两者有细微差异 一、 前期准备 创建一个工作目录 cd.. mkdir myslam && cd myslam 安装依赖 sudo apt-get update #安装git和cmake sudo apt-get install git cmake #安装opencv sudo apt-get install libopencv-dev #安装glew sudo apt-get install libgl
Ubuntu18.04 orbslam2环境配置
nonobodyyy的博客
04-29 245
把下载解压好的文件放入主目录，执行安装和编译。一键安装ROS自带opencv3够用。出现三色正方体则成功。
VLM-AD：通过视觉语言模型监督实现端到端自动驾驶
qq_54556560的博客
05-07 587
现有的端到端（E2E）自动驾驶（AD）模型通常经过优化，以模拟数据中观察到的驾驶模式，而不捕获底层推理过程。为了缩小这一差距，我们提出了VLM-AD，这是一种利用视觉语言模型（VLM）作为教师的方法，通过提供包含非结构化推理信息和结构化动作标签的额外监督来加强培训。而直接集成大语言模型（LLM）或视觉语言模型（VLM）的方法则需要大量微调，且推理时依赖大模型，计算开销大。现有的端到端（E2E）自动驾驶模型通过模仿数据中的驾驶模式进行优化，但缺乏人类驾驶员在复杂场景中的。”）生成非结构化文本。
[创业之路-350]：光刻机、激光器、自动驾驶、具身智能：跨学科技术体系全景解析（光-机-电-材-热-信-控-软-网-算-智）
文火冰糖（王文兵）的博客
05-02 969
光刻机、激光器、自动驾驶、具身智能四大领域的技术突破均依赖光、机、电、材、热、信、控、软、网、算、智十一大学科体系的深度耦合。的深度融合将催生万亿级新市场，而掌握这一技术体系的工程师将成为推动第四次工业革命的核心力量。现代技术系统的核心竞争力源于。
基于DQN的自动驾驶小车绕圈任务
weixin_44558122的博客
05-02 1166
任务来源: DQN: Deep Q Learning ｜自动驾驶入门（？） ｜算法与实现 任务原始代码: self-driving car 最终效果： 以下所有内容，都是对上面DQN代码的改进，在与deepseek的不断提问交互中学习理解代码，并且对代码做出改进，让小车能够尽可能的跑的远。 这个DQN代码依然有很多可改进的地方，受限于个人精力，没有再继续探索下去，欢迎指正交流。 后续会更新DDPG算法和PPO算法在此任务上的应用
Python赋能自动驾驶：如何优化路径规划，让AI驾驶更聪明？
Echo_Wish
05-07 157
从特斯拉的FSD，到Waymo的无人驾驶出租车，算法的精细化决定了自动驾驶的安全性与效率。是不可或缺的一环——如何在交通环境复杂多变的情况下，找到最优路径，并做出实时决策，是自动驾驶系统智能化的重要体现。下面，我们结合具体代码，看看如何用Python优化自动驾驶路径。，因此仅靠传统搜索算法还不够，我们可以利用。*，同时考虑了路径成本，确保驾驶系统选择。路径优化不仅仅是寻找最短路线，还涉及。，用于计算最短路径，同时考虑了。近年来，自动驾驶路径优化结合了。实现自动驾驶路径优化，并结合。，让算法更高效、更智能。
自动驾驶-一位从业两年的独特视角
m0_52186223的博客
04-30 772
2023.03 作为一名大三学生，加入到某量产车企，从事地图匹配研发2023.07 地图匹配项目交付，参与离线云端建图研发2023.10 拿到24届校招offer2024.07 正式入职2025.01 离线云端建图稳定，开始接触在线车端融图研发。
SafeDrive：大语言模型实现自动驾驶汽车知识驱动和数据驱动的风险-敏感决策——论文阅读
最新发布
qq_54556560的博客
05-07 531
为了解决这些问题，我们提出了SafeDrive，这是一个知识和数据驱动的风险敏感决策框架，以提高自动驾驶汽车的安全性和适应性。拟议框架引入了一个模块化系统，包括：（1）风险模块，用于全面量化涉及驾驶员、车辆和道路相互作用的多因素耦合风险；对以动态和高风险场景为特征的现实世界交通数据集进行广泛评估，包括高速公路（HighD）、十字路口（InD）和环形交叉口（RounD），验证了该框架提高决策安全性（实现100%的安全率）、复制类人驾驶行为（决策一致性超过85%）以及有效适应不可预测场景的能力。
LeapVAD：通过认知感知和 Dual-Process 思维实现自动驾驶飞跃——论文阅读
qq_54556560的博客
05-07 696
现有自动驾驶技术中，数据驱动方法依赖大量标注数据且缺乏复杂场景下的推理能力，而知识驱动方法（如基于视觉语言模型VLM）虽具备一定推理能力，但评估方法多为开环测试，无法反映动态交互环境。框架，融合认知感知（人类注意力机制）与双过程思维（分析过程System-II + 启发式过程System-I），结合记忆库和反射机制，实现闭环环境下的持续优化。记忆库（CARLA训练）跨域迁移有效，ADS（驾驶分数）达45.52%，优于端到端方法（如VAD、UniAD）。
自然语言处理实战：用CRF打造高精度命名实体识别系统
Loving_enjoy的博客
05-03 1393
无论是想快速搭建一个可用的NER系统，还是希望深入理解概率图模型的精髓，CRF都是值得放入工具箱的利器。'prev_is_b-geo': prev_tag == 'B-GEO' # 假设prev_tag是前一个标签。('参观天安门', ['O', 'B-POI', 'I-POI', 'I-POI'])" —— 人工智能先驱吴恩达。- **BiLSTM-CRF**：经典组合，在CoNLL-2003达到91%的F1值。('北京市', ['B-GEO', 'I-GEO', 'I-GEO']),
解锁DeepSeek模型微调：从小白到高手的进阶之路
邓邓子的博客
05-05 1532
本文围绕 DeepSeek 模型微调展开系统阐述。首先介绍 DeepSeek 模型在 AI 领域的重要地位及其优势，点明微调对提升模型性能的关键意义。接着深入解析微调原理，涵盖迁移学习基础与参数更新机制。随后详细讲解数据准备、模型选择加载、微调训练实战等核心步骤，包括数据收集标注预处理、参数设置与策略选择。还通过实战案例展示微调全流程，并基于评估结果提出优化改进方法。最后对 DeepSeek 模型微调进行总结，展望其未来发展方向，为希望掌握 DeepSeek 模型微调技术的读者提供全面指导。
ubantu18.04配置move_base
04-03
<think>好的，我现在需要在Ubuntu 18.04上配置ROS的move_base导航栈。首先，我需要了解用户的需求。用户使用的是Ubuntu 18.04和ROS Melodic版本，因为根据引用[1]和引用[3]提到的ros-melodic-navigation*和TurtleBot3-Burger的配置，都是基于Melodic的。所以，首先确认ROS Melodic已经正确安装。 接下来，用户需要配置move_base导航栈。根据引用[1]中的信息，安装move_base可以通过sudo apt install ros-melodic-navigation*命令来安装导航相关的所有包，包括move_base。但引用[4]提到可能存在ros-melodic-move-base-to-manip包不存在的情况，但正常编译move_base不会有异常，所以可能需要单独安装某些依赖。 然后，我需要检查是否有依赖需要安装。引用[2]提到安装g++、openssl和libssl-dev，这些都是编译和运行所需的依赖项。所以在安装move_base之前，应该先安装这些依赖。 步骤应该是：首先安装ROS Melodic，如果尚未安装的话。然后更新系统包，接着安装导航栈和相关依赖。可能还需要创建工作空间，并编译相关包，特别是如果用户需要自定义配置或使用特定机器人（如TurtleBot3），如引用[3]中的例子，可能需要额外的步骤来配置机器人模型和仿真环境。 另外，用户可能遇到的问题是依赖缺失，比如某些包无法找到，这时候需要根据错误信息安装对应的ROS包。例如，如果move_base包安装失败，可能需要单独安装ros-melodic-move-base，或者检查是否添加了正确的软件源。 测试配置是否成功可以通过运行示例，比如使用TurtleBot3的仿真环境，加载地图并测试导航功能。这需要安装TurtleBot3的包，可能包括仿真和导航相关的依赖。 总结步骤应该是：安装依赖 -> 安装导航栈 -> 创建工作区（可选）-> 配置机器人模型 -> 测试导航功能。过程中需要参考用户提供的引用内容，确保所有步骤正确，并处理可能出现的异常情况，如某些包不存在的情况。</think>### 分步配置指南 #### 一、安装依赖与环境准备 1. **更新系统软件包** 确保系统环境最新： ```bash sudo apt update && sudo apt upgrade -y ``` 2. **安装编译工具与基础依赖** 根据引用[2]，需安装编译工具链和加密库： ```bash sudo apt-get install g++ openssl libssl-dev ``` 3. **安装ROS Melodic（若未安装）** 若未安装ROS，需先执行完整安装： ```bash sudo sh -c &#39;echo "deb http://packages.ros.org/ros/ubuntu $(lsb_release -sc) main" > /etc/apt/sources.list.d/ros-latest.list&#39; sudo apt-key adv --keyserver &#39;hkp://keyserver.ubuntu.com:80&#39; --recv-key C1CF6E31E6BADE8868B172B4F42ED6FBAB17C654 sudo apt update sudo apt install ros-melodic-desktop-full echo "source /opt/ros/melodic/setup.bash" >> ~/.bashrc source ~/.bashrc ``` --- #### 二、安装导航栈核心组件 1. **安装move_base及相关导航包** 根据引用[1]和[4]，执行以下命令安装完整导航栈： ```bash sudo apt install ros-melodic-navigation* ``` 若出现`ros-melodic-move-base-to-manip`缺失（如引用[4]所述），可跳过或手动安装关键包： ```bash sudo apt install ros-melodic-move-base ros-melodic-amcl ros-melodic-map-server ``` 2. **验证安装** 检查`move_base`包是否存在： ```bash rospack find move_base ``` 若返回路径`/opt/ros/melodic/share/move_base`，则安装成功。 --- #### 三、配置机器人模型与仿真环境（以TurtleBot3为例） 1. **安装TurtleBot3依赖** 根据引用[3]，需安装仿真与导航支持包： ```bash sudo apt install ros-melodic-turtlebot3* ros-melodic-gazebo-ros-pkgs ``` 2. **设置默认机器人模型** 在`~/.bashrc`中添加环境变量： ```bash echo "export TURTLEBOT3_MODEL=burger" >> ~/.bashrc source ~/.bashrc ``` --- #### 四、运行导航功能测试 1. **启动Gazebo仿真环境** ```bash roslaunch turtlebot3_gazebo turtlebot3_world.launch ``` 2. **加载地图并启动导航节点** 新终端中执行： ```bash roslaunch turtlebot3_navigation turtlebot3_navigation.launch ``` 3. **通过Rviz设置导航目标** 在Rviz界面中点击 **2D Nav Goal** 并指定目标点，观察机器人自动规划路径。 --- §§ 1. 如何解决`move_base`启动时提示`global_costmap`参数错误？ *可能需检查`costmap_common_params.yaml`中的传感器配置[^4]* 2. TurtleBot3导航时无法避开障碍物如何调试？ *需验证激光雷达数据是否正常，并调整`local_costmap`的分层权重[^3]* 3. 如何将`move_base`集成到自定义机器人？ *需重新配置底盘控制、传感器URDF与导航参数文件[^1]* : 引用[1]: ubuntu18.04安装cartographer和move_base。二.安装move_base sudo apt install ros-melodic-navigation* [^2]: 引用[2]: 虚拟机环境ubuntu18.04 ros-melodic配置。安装依赖 sudo apt-get install g++ sudo apt-get install openssl sudo apt-get install libssl-dev [^3]: 引用[3]: TurtleBot3-Burger+Ubuntu18.04+Melodic ROS 安装记录。 : 引用[4]: ROS中move_base,navigation的软件包。这里面第一条最后一个 ros-melodic-move-base-to-manip 可能会不存在,但是小编正常编译move_base也没发现异常.
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值