Ubuntu16.04 + ROS Kinetic: SVO (data set running)

Ubuntu16.04 + ROS Kinetic: SVO (data set running)

Preparation：

• Install Ubuntu 16.04
• Build ROS Kinetic environment

main content:

• Install the library
• Run SVO with dataset

Install dependencies process:

• 1. Install Sophis (this library is a well-maintained Lie group and Lie algebra library)

#catkin_ws 是自己创建的ROS工作空间，在之前的博客里面也有创建ROS工作空间教程
cd catkin_ws/src/SVO  #后面源代码拷贝到这个目录下，所以将svo需要的库也放在这里面
git clone https://github.com/strasdat/Sophus.git
cd Sophus
git checkout a621ff
mkdir build
cd build
cmake ..
make

• 2. Install Fast (a library for Fast corner extraction)

cd catkin_ws/src/SVO
git clone https://github.com/uzh-rpg/fast.git
cd fast
mkdir build
cd build
cmake ..
make

• 3. Install g2o (a nonlinear optimization library, also like sophus, you have already installed it with high probability, optional, but it is recommended to install, as for environment variables, in fact, no need to set, just execute the following command)

cd catkin_ws/src/SVO
git clone https://github.com/RainerKuemmerle/g2o.git
cd g2o
mkdir build
cd build
cmake ..
make
sudo make install

• 4. Install vikit (this is actually a visual development kit, vision kit, especially pcl is very dependent on this library, it is strongly recommended to install. But sometimes different versions adapt to different slam systems. Svo can use the following installation method)

cd catkin_ws/src/SVO
git clone https://github.com/uzh-rpg/rpg_vikit.git

• 5. ROS dependencies (mainly cmake-modules, the version of Ubuntu I use is Ubuntu16.04 + ros-kinetic) I have not installed this dependency before, a lot of cmake-modules problems have occurred, and the solution is very rough, Cannot find cmake-modules, then I went to the corresponding CMakeLists.txt to delete the dependency on cmake-modules, and the result was that the compilation and operation was successful. ) But just in case, it is recommended to install the following dependencies.

sudo apt-get install ros-kinetic-cmake-modules

SVO run dataset

• 1. The first is the data set download. It is strongly recommended to download the data set by segmentation. The data set is large, about 1.7G, and the segmentation resume is very fast:

wget -c rpg.ifi.uzh.ch/datasets/airground_rig_s3_2013-03-18_21-38-48.bag
## 下载到catkin_ws目录下就可以

• 2. Open a new terminal

cd catkin_ws
source devel/setup.bash
roslaunch /home/<用户名>/catkin_ws/src/SVO/rpg_svo/svo_ros/launch/test_rig3.launch

• 3. Open a new terminal

rosrun rviz rviz -d /home/<用户名>/catkin_ws/src/SVO/rpg_svo/svo_ros/rviz_config.rviz

• 4. Open a new terminal and run the data set

cd catkin_ws
rosbag play airground_rig_s3_2013-03-18_21-38-48.bag 

running result

specific method

Reference blog original address: https://blog.csdn.net/ouyangandy/article/details/89634871