自主移动机器人（2WD、4WD、MecanumDrive）_Python_Shell_下载.zip

# linorobot2  linorobot2 is a ROS2 port of the [linorobot](https://github.com/linorobot/linorobot) package. If you're planning to build your own custom ROS2 robot (2WD, 4WD, Mecanum Drive) using accessible parts, then this package is for you. This repository contains launch files to easily integrate your DIY robot with Nav2 and a simulation pipeline to run and verify your experiments on a virtual robot in Gazebo. Once the robot's URDF has been configured in linorobot2_description package, users can easily switch between booting up the physical robot and spawning the virtual robot in Gazebo.  Assuming you're using one of the tested sensors, linorobot2 automatically launches the necessary hardware drivers, with the topics being conveniently matched with the topics available in Gazebo. This allows users to define parameters for high level applications (ie. Nav2 SlamToolbox, AMCL) that are common to both virtual and physical robots. The image below summarizes the topics available after running **bringup.launch.py**.  An in-depth tutorial on how to build the robot is available in [linorobot2_hardware](https://github.com/linorobot/linorobot2_hardware). ## Installation This package requires ros-foxy or ros-galactic. If you haven't installed ROS2 yet, you can use this [installer](https://github.com/linorobot/ros2me) script that has been tested to work on x86 and ARM based dev boards ie. Raspberry Pi4/Nvidia Jetson Series. ### 1. Robot Computer - linorobot2 Package The easiest way to install this package on the robot computer is to run the bash script found in this package's root directory. It will install all the dependencies, set the ENV variables for the robot base and sensors, and create a linorobot2_ws (robot_computer_ws) on the robot computer's `$HOME` directory. If you're using a ZED camera with a Jetson Nano, you must create a custom Ubuntu 20.04 image for CUDA and the GPU driver to work. Here's a quick [guide](./ROBOT_INSTALLATION.md#1-creating-jetson-nano-image) on how to create a custom image for Jetson Nano. source /opt/ros/<ros_distro>/setup.bash cd /tmp wget https://raw.githubusercontent.com/linorobot/linorobot2/${ROS_DISTRO}/install_linorobot2.bash bash install_linorobot2.bash <robot_type> <laser_sensor> <depth_sensor> source ~/.bashrc robot_type: - `2wd` - 2 wheel drive robot. - `4wd` - 4 wheel drive robot. - `mecanum` - Mecanum drive robot. laser_sensor: - `rplidar` - [RP LIDAR A1](https://www.slamtec.com/en/Lidar/A1) - `ldlidar` - [LD06 LIDAR](https://www.inno-maker.com/product/lidar-ld06/) - `ydlidar` - [YDLIDAR](https://www.ydlidar.com/lidars.html) - `xv11` - [XV11](http://xv11hacking.rohbotics.com/mainSpace/home.html) - `realsense` - * [Intel RealSense](https://www.intelrealsense.com/stereo-depth/) D435, D435i - `zed` - * [Zed](https://www.stereolabs.com/zed) - `zed2` - * [Zed 2](https://www.stereolabs.com/zed-2) - `zed2i` - * [Zed 2i](https://www.stereolabs.com/zed-2i) - `zedm` - * [Zed Mini](https://www.stereolabs.com/zed-mini) - `-` - If the robot's sensor is not listed above. Sensors marked with an asterisk are depth sensors. If a depth sensor is used as a laser sensor, the launch files will run [depthimage_to_laserscan](https://github.com/ros-perception/depthimage_to_laserscan) to convert the depth sensor's depth image to laser scans. depth_sensor: - `realsense` - [Intel RealSense](https://www.intelrealsense.com/stereo-depth/) D435, D435i - `zed` - [Zed](https://www.stereolabs.com/zed) - `zed2` - [Zed 2](https://www.stereolabs.com/zed-2) - `zed2i` - [Zed 2i](https://www.stereolabs.com/zed-2i) - `zedm` - [Zed Mini](https://www.stereolabs.com/zed-mini) Alternatively, follow this [guide](./ROBOT_INSTALLATION.md) to do the installation manually. ### 2. Host Machine / Development Computer - Gazebo Simulation (Optional) This step is only required if you plan to use Gazebo later. This comes in handy if you want to fine-tune parameters (ie. SLAM Toolbox, AMCL, Nav2) or test your applications on a virtual robot. #### 2.1 Install linorobot2 Package Install linorobot2 package on the host machine: cd <host_machine_ws> git clone -b $ROS_DISTRO https://github.com/linorobot/linorobot2 src/linorobot2 rosdep update && rosdep install --from-path src --ignore-src -y --skip-keys microxrcedds_agent --skip-keys micro_ros_agent colcon build source install/setup.bash * microxrcedds_agent and micro_ros_agent dependency checks are skipped to prevent this [issue](https://github.com/micro-ROS/micro_ros_setup/issues/138) of finding its keys. This means that you have to always add `--skip-keys microxrcedds_agent --skip-keys micro_ros_agent` whenever you have to run `rosdep install` on the ROS2 workspace where you installed linorobot2. #### 2.2 Define Robot Type Set LINOROBOT2_BASE env variable to the type of robot base used. Available env variables are *2wd*, *4wd*, and *mecanum*. For example: echo "export LINOROBOT2_BASE=2wd" >> ~/.bashrc source ~/.bashrc You can skip the next step (Host Machine - RVIZ Configurations) since this package already contains the same RVIZ configurations to visualize the robot. ### 3. Host Machine - RVIZ Configurations Install [linorobot2_viz](https://github.com/linorobot/linorobot2_viz) package to visualize the robot remotely specifically when creating a map or initializing/sending goal poses to the robot. The package has been separated to minimize the installation required if you're not using the simulation tools on the host machine. cd <host_machine_ws> git clone https://github.com/linorobot/linorobot2_viz src/linorobot2_viz rosdep update && rosdep install --from-path src --ignore-src -y colcon build source install/setup.bash ## Hardware and Robot Firmware All the hardware documentation and robot microcontroller's firmware can be found [here](https://github.com/linorobot/linorobot2_hardware). ## URDF ### 1. Define robot properties [linorobot2_description](./linorobot2_description) package has parameterized xacro files that can help you kickstart writing the robot's URDF. Open <robot_type>.properties.urdf.xacro in [linorobot2_description/urdf](./linorobot2_description/urdf) directory and change the values according to the robot's specification/dimensions. All pose definitions must be measured from the `base_link` (center of base) and wheel positions (ie `wheel_pos_x`) are referring to wheel 1. For custom URDFs, you can change the `urdf_path` in [description.launch.py](./linorobot2_description/launch/description.launch.py) found in linorobot2_description/launch directory. Robot Orientation: --------------FRONT-------------- WHEEL1 WHEEL2 (2WD/4WD) WHEEL3 WHEEL4 (4WD) --------------BACK-------------- Build the robot computer's workspace to load the new URDF: cd <robot_computer_ws> colcon build The same changes must be made on the host machine's <robot_type>.properties.urdf.xacro if you're simulating the robot in Gazebo. Remember to also build the host machine's workspace after editing the xacro file. cd <host_machine_ws> colcon build ### 2. Visualize the newly created URDF #### 2.1 Publish the URDF from the robot computer: ros2 launch linorobot2_description description.launch.py Optional parameters for simulation on host machine: - **rviz** - Set to true to visualize the robot in rviz2 and only if you're configuring the URDF from the host machine. For example: ros2 launch linorobot2_description description.launch.py rviz:=true #### 2.2 Visualize the robot from the host machine: The `rviz` argument on description.launch.py won't work on headless setup but you can visualize the robot remotely from the host machine: ros2 launch linorobot2_viz robot_model.launch.py ## Quickstart All commands below are to be run on the robot