If it's your first time using ROS2 and haven't created your ROS2 workspace yet, you can check out ROS2 Creating a Workspace tutorial.
source /opt/ros/<your_ros_distro>/setup.bash
cd <your_ws>
colcon build
source install/setup.bash
RPLIDAR:
sudo apt install -y ros-$ROS_DISTRO-rplidar-ros
cd /tmp
wget https://raw.githubusercontent.com/allenh1/rplidar_ros/ros2/scripts/rplidar.rules
sudo cp rplidar.rules /etc/udev/rules.d/
LDLIDAR:
cd <your_ws>
git clone https://github.com/linorobot/ldlidar src/ldlidar
sudo cp src/ldlidar/ldlidar.rules /etc/udev/rules.d/
colcon build
source <your_ws>/install/setup.bash
YDLIDAR:
cd /tmp
git clone https://github.com/YDLIDAR/YDLidar-SDK.git
cd YDLidar-SDK/build
cmake ..
make
sudo make install
cd <your_ws>
git clone https://github.com/YDLIDAR/ydlidar_ros2_driver src/ydlidar_ros2_driver
chmod 0777 src/ydlidar_ros2_driver/startup/*
sudo sh src/ydlidar_ros2_driver/startup/initenv.sh
colcon build --symlink-install
source <your_ws>/install/setup.bash
XV11:
cd <your_ws>
git clone https://github.com/mjstn/xv_11_driver src/xv_11_driver
colcon build
source <your_ws>/install/setup.bash
Intel RealSense:
sudo apt install ros-$ROS_DISTRO-realsense2-camera
Zed Camera:
cd /tmp
wget https://download.stereolabs.com/zedsdk/3.5/jp45/jetsons -O zed_sdk #use Jetson SDK
#wget https://download.stereolabs.com/zedsdk/3.5/cu111/ubuntu20 -O zed_sdk #use this for x86 machine with NVIDIA GPU
chmod +x zed_sdk
./zed_sdk -- silent
cd <your_ws>
git clone https://github.com/stereolabs/zed-ros2-wrapper src/zed-ros2-wrapper
git clone https://github.com/ros-perception/image_common -b $ROS_DISTRO src/image_common
rosdep install --from-paths src --ignore-src -r -y
colcon build --symlink-install --cmake-args=-DCMAKE_BUILD_TYPE=Release
source <your_ws>/install/setup.bash
cd <your_ws>
git clone -b $ROS_DISTRO https://github.com/micro-ROS/micro_ros_setup src/micro_ros_setup
sudo apt install python3-vcstool build-essential
sudo apt update && rosdep update
rosdep install --from-path src --ignore-src -y
colcon build
source install/setup.bash
ros2 run micro_ros_setup create_agent_ws.sh
ros2 run micro_ros_setup build_agent.sh
source install/setup.bash
- You can ignore
1 package had stderr output: microxrcedds_agent
after building your workspace.
cd <your_ws>
git clone -b $ROS_DISTRO https://github.com/linorobot/linorobot2 src/linorobot2
If you're installing this on the robot's computer or you don't need to run Gazebo at all, you can skip linorobot2_gazebo package by creating a COLCON_IGNORE file:
cd src/linorobot2/linorobot2_gazebo
touch COLCON_IGNORE
cd <your_ws>
rosdep update && rosdep install --from-path src --ignore-src -y --skip-keys microxrcedds_agent
colcon build
source install/setup.bash
- microxrcedds_agent dependency checks are skipped to prevent this issue of finding its keys. This means that you have to always add
--skip-keys microxrcedds_agent
whenever you have to runrosdep install
on the ROS2 workspace where you installed linorobot2.
Set LINOROBOT2_BASE env variable to the type of robot base that you want to use. This is not required if you're using a custom URDF. Available env variables are 2wd, 4wd, and mecanum. For example:
echo "export LINOROBOT2_BASE=2wd" >> ~/.bashrc
The Nav2 config file has been configured to support Voxel Layer for marking 3D obstacles in the Local Costmap using a depth sensor. To enable one of the tested depth sensor's launch files in bringup.launch.py, export the depth sensor you're using to LINOROBOT2_DEPTH_SENSOR
env variable.
Tested sensors are:
realsense
- Intel RealSense D435, D435iastra
- Orbec Astrazed
- Zedzed2
- Zed 2zed2i
- Zed 2izedm
- Zed Mini
For example:
echo "export LINOROBOT2_DEPTH_SENSOR=realsense" >> ~/.bashrc
The launch files of the tested laser sensors have already been added in bringup.launch.py. You can enable one of these sensors by exporting the laser sensor you're using to LINOROBOT2_LASER_SENSOR
env variable.
Tested Laser Sensors:
rplidar
- RP LIDAR A1ldlidar
- LD06 LIDARydlidar
- YDLIDARrealsense
- Intel RealSense D435, D435iastra
- Orbec Astrazed
- Zedzed2
- Zed 2zed2i
- Zed 2izedm
- Zed Mini
For example:
echo "export LINOROBOT2_LASER_SENSOR=rplidar" >> ~/.bashrc
If you export a depth sensor to LINOROBOT2_LASER_SENSOR
, the launch file will run depthimage_to_laserscan to convert the depth sensor's depth image to laser.
Source your ~/.bashrc
to apply the changes you made:
source ~/.bashrc
-
Create the custom image on your host machine by following this tutorial. Prepare an SD card and use the helper scripts found in the repository mentioned in the tutorial to build and flash the custom image to the SD card.
-
Before going through the tutorial, you can change the user name and password by modifying this file.
-
If you encounter this problem:
Extend the fs... e2fsck: No such file or directory while trying to open /dev/sdcp1
after flashing the image, you can check out this issue to resolve the problem. -
On your jetson nano, install the Nvidia Libraries:
sudo apt install -y cuda-toolkit-10-2 libcudnn8 libcudnn8-dev
This is a short tutorial on how to make your bringup launch files run during startup.
sudo touch /etc/ros/env.sh
sudo nano /etc/ros/env.sh
and paste the following:
#!/bin/sh
export LINOROBOT2_BASE=<your_robot_type>
export LINOROBOT2_LASER_SENSOR=<your_supported_sensor> #(optional)
sudo touch /etc/systemd/system/robot-boot.service
sudo nano /etc/systemd/system/robot-boot.service
and paste the following:
[Unit]
After=NetworkManager.service time-sync.target
[Service]
Type=simple
User=<user>
ExecStart=/bin/sh -c ". /opt/ros/<your_ros_distro>/setup.sh;. /etc/ros/env.sh;. /home/<user>/<your_ws>/install/setup.sh; ros2 launch linorobot2_bringup bringup.launch.py joy:=true"
[Install]
WantedBy=multi-user.target
Remember to replace:
user
with your machine's user name (echo $USER
)your_ros_distro
with the ros2 distro (echo $ROS_DISTRO
) your machine is running onyour_ws
with the location of the ros2 ws where you installed linorobot2
sudo systemctl enable robot-boot.service
You can check if the service you just created is correct by:
sudo systemctl start robot-boot.service
sudo systemctl status robot-boot.service
- You should see the ros2 logs that you usually see when running bringup.launch.py. Once successful, you can now reboot your machine. bringup.launch.py should start running once the machine finished booting up.
systemctl stop robot-boot.service
systemctl disable robot-boot.service
sudo rm /etc/systemd/system/robot-boot.service
Source: https://blog.roverrobotics.com/how-to-run-ros-on-startup-bootup/