English | 中文 / Chinese

This project is an integrated control and teaching system for the WLKATA MT4 series robotic arm, mainly including:

• Robot model and description (wlkata_mt4_description): URDF/Xacro, joint configuration, etc., used to display the robot in RViz / simulation.
• MoveIt motion planning configuration (wlkata_mt4_moveit_config): MoveIt 2 configuration package for MT4 path planning and visualization.
• Real robot control node (wlkata_arm_move): Subscribes to trajectories planned by MoveIt, converts them to G-code, and sends them to the physical robot via serial port.
• Python control SDK (wlkatapython): A standalone Python package that controls Mirobot, E4, MT4, MS4220 and other devices via serial + G-code.

Typical use cases:

• Use ROS 2 + MoveIt for MT4 motion planning and visualization.
• Seamlessly send MoveIt planned trajectories to the real robot.
• Perform secondary development / upper-computer programming based on wlkatapython.

Key project directories:

• wlkata_mt4_description/

  • urdf/: wlkata_mt4_description.urdf, .xacro and other robot model files.
  • config/: YAML files for joint names and configurations.
  • launch/display.launch.py: Display the MT4 model in RViz.

• wlkata_mt4_moveit_config/

  • config/: MoveIt configuration files such as kinematics.yaml, ros2_controllers.yaml, wlkata_mt4_description.srdf, etc.
  • launch/:
    • demo.launch.py: Common MoveIt + RViz demo entry.
    • move_group.launch.py, moveit_rviz.launch.py, rsp.launch.py, etc.
  • package.xml: MoveIt configuration package definition.

• wlkata_arm_move/

  • wlkata_arm_move/mt4_moveit_move.py: Core Python node that subscribes to /display_planned_path (moveit_msgs/DisplayTrajectory), converts the planned joint trajectory into G-code commands, and sends them to the robot via serial port.
  • package.xml: ROS 2 Python package definition, depending on rclpy, moveit_msgs, wlkata_interfaces, action_msgs, std_msgs, etc.
  • setup.py: Exposes the executable node mt4_moveit_move via entry_points.

• wlkatapython/

  • wlkatapython.py and multiple subdirectories (Mirobot_robot/, E4_robot/, MT4_robot/, MS4220_robot/, Harobot_robot/, etc.): Serial control implementation for various devices.
  • README.md: Detailed introduction on how to control MT4/E4/MS4220 and other devices using Python + serial + G-code (including sample code and wiring diagrams).
  • setup.py: Packaging configuration of the standalone Python package, which can be used by installing with pip.

• Operating system: Ubuntu 22.04 / 24.04 that supports ROS 2 + MoveIt 2 is recommended.
• ROS 2 + MoveIt 2: Includes dependencies such as ament_cmake, moveit_ros_move_group, moveit_ros_visualization, robot_state_publisher, rviz2, etc. (see each package.xml).
• Python environment: Python ≥ 3.9.
• Build tools: colcon.

Recommended ROS 2 workspace layout (example):

~/ros2_ws/
  ├─ src/
  │   ├─ wlkata_mt4_description/
  │   ├─ wlkata_mt4_moveit_config/
  │   ├─ wlkata_arm_move/
  │   └─ wlkatapython/
  └─ ...

• Where to place this repository: Put the whole project folder under ~/ros2_ws/src/ (so that you can see the above packages directly under src/).
• Where to run commands:
  • sudo apt install ...: Can be run from any directory.
  • cd ~/ros2_ws: Go to the ROS 2 workspace root.
  • colcon build: Run in ~/ros2_ws.
  • source install/setup.bash: Run in ~/ros2_ws or any other terminal to load the workspace environment.
  • ros2 launch ... / ros2 run ...: Run in a terminal where source install/setup.bash has already been executed (current directory does not matter).

sudo apt install \
  ros-jazzy-joint-state-publisher-gui \
  ros-jazzy-moveit \
  ros-jazzy-controller-manager \
  ros-jazzy-joint-trajectory-controller \
  ros-jazzy-joint-state-broadcaster

cd ~/ros2_ws          # replace with your workspace
colcon build
source install/setup.bash

ros2 launch wlkata_mt4_description display.launch.py
ros2 launch wlkata_mt4_moveit_config demo.launch.py

• ros2 launch wlkata_mt4_description display.launch.py: Only launches the robot description and RViz. You can see the MT4 model, joint state publisher/GUI, etc. Mainly used to check whether URDF and joint configuration are correct. It does not include the MoveIt planning interface.
• ros2 launch wlkata_mt4_moveit_config demo.launch.py: Launches the MoveIt + RViz demo scene. The MotionPlanning panel will appear on the left side in RViz, and you can interactively plan and execute trajectories (in simulation or together with the real robot node).

Package names are basically the same, just replace the ros-jazzy-* prefix with ros-humble-*:

sudo apt install \
  ros-humble-joint-state-publisher-gui \
  ros-humble-moveit \
  ros-humble-controller-manager \
  ros-humble-joint-trajectory-controller \
  ros-humble-joint-state-broadcaster

cd ~/ros2_ws          # replace with your workspace
colcon build
source install/setup.bash

ros2 launch wlkata_mt4_description display.launch.py
ros2 launch wlkata_mt4_moveit_config demo.launch.py

• Ubuntu 24.04 + ROS 2 Jazzy: Core dependencies can be installed via the apt command above. For development, you may additionally install python3-colcon-common-extensions, python3-rosdep, etc.
• Ubuntu 22.04 + ROS 2 Humble: Replace all ros-jazzy-* package names with ros-humble-*. If ROS 2 is not yet installed, it is recommended to first install the desktop version following the official ROS documentation.

The serial communication part mainly refers to wlkatapython/README.md:

• Use G-code protocol + serial communication (RS485/UART) to control the robot arm, linear slide, conveyor belt and other devices.
• Usually you need a multi-function controller to fully use all features. If the robot is connected directly via serial port, some functions may not be available.

Example (UART mode, MT4 homing):

import wlkatapython
import serial

Serial1 = serial.Serial("/dev/ttyUSB0", 115200)
mt41 = wlkatapython.MT4_UART()
mt41.init(Serial1, -1)
mt41.homing()
Serial1.close()

In mt4_moveit_move.py, the default serial device is:

serial.Serial("/dev/ttyUSB0", 115200, timeout=1)

• Please confirm the actual serial port name (typically /dev/ttyUSB0, /dev/ttyUSB1, etc.) and modify according to your environment.

• Provides the MT4 robot model description (URDF/Xacro), joint configuration, and Launch files for display/simulation. Can be used to verify whether the model and coordinate frames are correct.

• Generated using MoveIt Setup Assistant. Contains kinematics, controllers, planning scene and other configurations. Provides entries such as demo.launch.py to start the MoveIt + RViz demo scene.

• ROS 2 Python package. The core node mt4_moveit_move receives trajectories planned by MoveIt, converts them into G-code that conforms to the WLKATA controller, sends them to the physical robot via serial port, and handles responses.

• Standalone Python package that can be used by installing with pip. Supports multiple devices including MT4, E4, MS4220, and supports both RS485 and UART communication. Its README.md contains many examples and wiring diagrams. The ROS 2 control nodes in this project directly reuse its communication logic.