Ubuntu 14.04 LTS+ ROS indigo+ORB-SLAM1

这两天把论文：ORB-SLAM: a Versatile and Accurate Monocular SLAM System中提供的代码跑了下.

出于赠人玫瑰，手有余香的理念，现在把运行的详细过程分享下：

1. 创建ROS工作空间

zhuquan@zhuquan-Precision-5510:~$ mkdir -p  ~/orbslam_catkin_ws/src
zhuquan@zhuquan-Precision-5510:~$ cd ~/orbslam_catkin_ws/src
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src$ catkin_init_workspace
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src$  cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$  catkin_make
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash

运行以下命令确认你的当前目录是否在环境变量中

zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ echo $ROS_PACKAGE_PATH

如果终端输出

/home/zhuquan/orbslam_catkin_ws/src:/opt/ros/indigo/share:/opt/ros/indigo/stacks

则说明ROS工作空间已经搭建好了。

2. Prerequisites (dependencies)

2.1 Boost
We use the Boost library to launch the different threads of our SLAM system.
sudo apt-get install libboost-all-dev 

2.2 ROS
We use ROS to receive images from the camera or from a recorded sequence (rosbag), and for visualization (rviz, image_view). 
**We have tested ORB-SLAM in Ubuntu 12.04 with ROS Fuerte, Groovy and Hydro; and in Ubuntu 14.04 with ROS Indigo**. 
If you do not have already installed ROS in your computer, we recommend you to install the Full-Desktop version of ROS Fuerte (http://wiki.ros.org/fuerte/Installation/Ubuntu).

2.3 OpenCV
We use OpenCV to manipulate images and features. If you use a ROS version older than ROS Indigo, OpenCV is already included in the ROS distribution. In newer version of ROS, OpenCV is not included and you will need to install it. **We tested OpenCV 2.4**. Dowload and install instructions can be found at: http://opencv.org/
2.4 g2o (included in Thirdparty)
We use a modified version of g2o (see original at https://github.com/RainerKuemmerle/g2o) to perform optimizations.
In order to compile g2o you will need to have installed Eigen3 (at least 3.1.0).
sudo apt-get install libeigen3-dev
2.5 DBoW2 (included in Thirdparty)
We make use of some components of the DBoW2 and DLib library (see original at https://github.com/dorian3d/DBoW2) for place recognition and feature matching. There are no additional dependencies to compile DBoW2.

另外blas和lapack安装如下：

sudo apt-get install libblas-dev liblapack-dev

其中，opencv已安装，安装的是2.4.13版的。

3.Installation

3.1 Clone the repository

zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src$ git clone https://github.com/raulmur/ORB_SLAM.git ORB_SLAM

3.2 Add the path where you cloned ORB-SLAM to the `ROS_PACKAGE_PATH` environment variable.

To do this, modify your .bashrc (在/home/yourusername/下，按下Ctrl+h复合键就可以见到) and add at the bottom the following line (replace PATH_TO_PARENT_OF_ORB_SLAM):

export ROS_PACKAGE_PATH=${ROS_PACKAGE_PATH}:PATH_TO_PARENT_OF_ORB_SLAM

例如：

export ROS_PACKAGE_PATH=${ROS_PACKAGE_PATH}:/home/zhuquan/orbslam_catkin_ws/src/

3.3  Build g2o. 

Go into `Thirdparty/g2o/` and execute:

zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src$ cd ORB_SLAM/Thirdparty/g2o
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o$ mkdir build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o$ cd build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o/build$ cmake .. -DCMAKE_BUILD_TYPE=Release
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o/build$ make

3.4 Build DBoW2. 

Go into Thirdparty/DBoW2/ and execute:

zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o/build$ cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/g2o$ cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty$ cd DBoW2/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2$ mkdir build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2$ cd build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2/build$ cmake .. -DCMAKE_BUILD_TYPE=Release
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2/build$ make

3.5 Build ORB_SLAM.

 In the ORB_SLAM root execute:
 注意：**If you use ROS Indigo, remove the depency of opencv2 in the manifest.xml.**

zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2/build$ cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty/DBoW2$ cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Thirdparty$ cd ..
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ mkdir build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ cd build
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/build$ cmake .. -DROS_BUILD_TYPE=Release
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/build$ make

好了，到此ORB-SLAM就安装好了，下面就来跑它了。:-)

4. Usage1: 逐个手动启动各个程序

4.1 打开一个新终端，运行 

zhuquan@zhuquan-Precision-5510:~$ roscore

运行结果如下图所示：

4.2 重新打开一个新终端（原来的终端保留），运行：

  rosrun ORB_SLAM ORB_SLAM PATH_TO_VOCABULARY PATH_TO_SETTINGS_FILE

其中，PATH_TO_VOCABULARY和PATH_TO_SETTINGS_FILE分别是ORB vocabulary 和 settings file的路径，这里的路径必须是绝对路径或者是相对于ORB-SLAM源代码所在目录的相对路径。

例如，这里使用的是相对路径：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src$ cd ORB_SLAM/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ rosrun ORB_SLAM ORB_SLAM Data/ORBvoc.txt Data/Settings.yaml

运行结果如下图:

4.2 The last processed frame is published to the topic `/ORB_SLAM/Frame`. You can visualize it using `image_view`：

rosrun image_view image_view image:=/ORB_SLAM/Frame _autosize:=true；

例如：

重新打开一个新终端（原来的终端保留），运行：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ rosrun image_view image_view image:=/ORB_SLAM/Frame _autosize:=true

运行结果下图所示：

4.3 The map is published to the topic `/ORB_SLAM/Map`, the current camera pose and global world coordinate origin are sent through `/tf` in frames `/ORB_SLAM/Camera` and `/ORB_SLAM/World` respectively.  Run `rviz` to visualize the map

注意：

        *in ROS Fuerte*:
运行：

     rosrun rviz rviz -d Data/rviz.vcg

*in ROS Groovy or a newer version*:

运行：

     rosrun rviz rviz -d Data/rviz.rviz

例如：

重新打开一个新终端（原来的终端保留），运行：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src/ORB_SLAM/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ rosrun rviz rviz -d Data/rviz.rviz

运行结果如下图所示：

4.4 ORB_SLAM will receive the images from the topic `/camera/image_raw`. You can now play your rosbag。

5. Usage2: Use a roslaunch to launch `ORB_SLAM`, `image_view` and `rviz` from just one instruction.

We provide an example**:
*in ROS Fuerte*:

roslaunch ExampleFuerte.launch

*in ROS Groovy or a newer version*:

roslaunch ExampleGroovyOrNewer.launch

例如：

5.1 打开一个新终端，运行 

zhuquan@zhuquan-Precision-5510:~$ roscore

运行结果如下图所示：

5.2 重新打开一个终端（原终端保留），运行：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src/ORB_SLAM/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ roslaunch ExampleGroovyOrNewer.launch

运行结果如下图所示：

6. Example Sequence

We provide the settings and the rosbag of an example sequence in our lab. In this sequence you will see a loop closure and two relocalisation from a big viewpoint change.
6,1 Download the rosbag file and uncompress the file.  
http://webdiis.unizar.es/~raulmur/orbslam/downloads/Example.bag.tar.gz.

6.2 Once the ORB vocabulary has been loaded, play the rosbag (press space to start):
rosbag play --pause Example.bag

例如：

Usage1:

重新打开一个新终端（原来的终端保留），运行：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src/ORB_SLAM/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ cd Data
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Data$ rosbag play --pause Example.bag

运行结果如下图所示：

Usage2:

重新打开一个终端（原终端保留），运行：

zhuquan@zhuquan-Precision-5510:~$ cd orbslam_catkin_ws/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ source devel/setup.bash
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws$ cd src/ORB_SLAM/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM$ cd Data/
zhuquan@zhuquan-Precision-5510:~/orbslam_catkin_ws/src/ORB_SLAM/Data$ rosbag play --pause Example.bag

运行结果如下图所示：

到此，程序就跑完了～

主要参考：论文源代码中的Readme文档。