Laser Odometry and Mapping (Loam)
1. 동작 과정
구조와 소스코드를 분석해 보면 다음과 같은 순서로 SLAM데이터를 계산하고 있는 것을 알 수 있다.
- velodyne node: 스캔 후 point cloud 데이터 획득
- multi scan registration node: 스캔된 데이터에서 곡률 기반 특징점(모서리, 평면) 생성
- laser odometry: 이전 스캔 데이터에서 생성된 특징점을 기반으로 현재 스캔 데이터의 매특징점과 비교해 주행괘적(odometry)을 계산
- laser mapping: 주행괘적을 이용해 스캔 데이터를 보정해 정합함. 만약, IMU 센서가 있는 경우, /imu/data에서 얻은 데이터를 이용해 데이터 보정함
설치
OS : 16.04 and 14.04
PCL : 1.8 추천 (1.7은 multiScanRegistration error발생)
# PCL 1.8 설치 Kinetic
$ apt-get install git build-essential linux-libc-dev cmake cmake-gui libusb-1.0-0-dev libusb-dev libudev-dev mpi-default-dev openmpi-bin openmpi-common libflann1.8 libflann-dev libeigen3-dev libboost-all-dev libvtk5.10-qt4 libvtk5.10 libvtk5-dev libqhull* libgtest-dev freeglut3-dev pkg-config libxmu-dev libxi-dev mono-complete qt-sdk openjdk-8-jdk openjdk-8-jre
$ git clone https://github.com/PointCloudLibrary/pcl.git
$ cd pcl && mkdir release && cd release
$ cmake -DCMAKE_BUILD_TYPE=None -DCMAKE_INSTALL_PREFIX=/usr -DBUILD_GPU=ON -DBUILD_apps=ON -DBUILD_examples=ON -DCMAKE_INSTALL_PREFIX=/usr ..
$ make -j8
$ sudo make install
$ sudo apt-get install ros-kinetic-pcl-conversions ros-kinect-pcl-ros
# Loam compilation and installation
$ source /opt/ros/$ROS_DISTRO/setup.bash
$ cd ~
$ mkdir -p catkin_ws/src
$ cd catkin_ws/src
$ catkin_init_workspace
$ cd ~/catkin_ws/src
$ git clone https://github.com/laboshinl/loam_velodyne.git
## 설치전 사전 작업
$ vi src/lib/LaserMapping.cpp ## 주석 처리 139-153 #https://github.com/laboshinl/loam_velodyne/pull/84/files
$ cd ~/catkin_ws
$ rosdep install --from-paths ./src/loam_velodyne/
$ catkin_make #$ catkin_make -DCMAKE_BUILD_TYPE=Release
$ source ~/catkin_ws/devel/setup.bash
$ echo "source ~/catkin_ws/devel/setup.bash" >> ~/.bashrc
#Download package
$ wget http://www.aisl.cs.tut.ac.jp/databases/hdl_graph_slam/hdl_400.bag.tar.gz
$ tar -axvf hdl_400.bag.tar.gz
## NSH indoor outdoor : [[Download]](https://pan.baidu.com/s/18ISyr4ky2MfTl7TXJD2W-A), 비번 2yea
# Run LOAM
$ roslaunch loam_velodyne loam_velodyne.launch
$ rosbap play hdl_400.bag #rosbag play data_file_name.bag -r 0.5 (속도 줄이기)
Error handling
1.multiScanRegistration error
If the bag is prompted to multiScanRegistration error after running Loam, the reason is to use apt-get to install the compiled pcl. You need to uninstall PCL according to the second part and download the source code to compile and install PCL. Then you need to restart catkin_make.
2.error: 'downSizeFilterMap' was not declared in this scope error
Errors may appear in catkin_make: 'downSizeFilterMap' was not declared in this scope error, because the source code of Loam has not been modified, refer to https://github.com/laboshinl/loam_velodyne/pull/84/files for src/lib Comment out the 139-153 lines of /LaserMapping.cpp
Save and view Loam's 3D point cloud map
지도 정보는 /laser_cloud_surround토픽으로 출력.. 이를 수신하여 pcd로 저장
rosbag record -o out /laser_cloud_surround
rosrun pcl_ros bag_to_pcd input.bag /laser_cloud_surround pcd
# 필요시 ply로 변환
pcl_pcd2ply xxxx.pcd
센서 데이터 바로 적용 하기
Vlp16.yaml 준비
- 다운로드 :
wget https://raw.githubusercontent.com/Kitware/VeloView/master/share/VLP-16.xml - xml을 변환 :
$ rosrun velodyne_pointcloud gen_calibration.py ~/Desktop/VLP-16.xml
$ roslaunch velodyne_pointcloud VLP16_points.launch calibration:=/home/xxxx/VLP-16.yaml
$ roslaunch loam_velodyne loam_velodyne.launch
배경 저장 : rosrun pcl_ros pointcloud_to_pcd input:=/velodyne_cloud_registered