Ros MCP Server

🚀 ROS MCP Server

ROS MCP Server aims to support robots in performing complex tasks and effectively adapting to various environments. It provides a set of functions to convert natural - language commands input by users through LLM (Large Language Model) into ROS commands for robot control. Additionally, by using rosbridge, it is configured to work with both ROS and ROS2 systems. Its WebSocket - based communication makes it widely applicable across various platforms.

The currently supported message type is:

• geometry_msgs/Twist

📦 Installation

Install via Smithery

To automatically install ros-mcp-server for the Claude desktop environment, visit Smithery and use the following command:

npx -y @smithery/cli install @lpigeon/ros-mcp-server --client claude

Local Installation

Install using uv:

• To install uv, you can use the following command:

curl -LsSf https://astral.sh/uv/install.sh | sh

Or

pip install uv

• Create and activate a virtual environment (optional)

uv venv
source .venv/bin/activate

MCP Server Configuration:

Set up MCP in mcp.json.

"ros-mcp-server": {
  "command": "uv",
  "args": [
    "--directory",
    "/ABSOLUTE/PATH/TO/PARENT/FOLDER/ros-mcp-server",
    "run",
    "server.py"
  ]
}

✨ Features

pub_twist

• Purpose: Send motion commands to the robot by setting linear and angular velocities.
• Parameters:
  • linear: Linear velocity (List[Any])
  • angular: Angular velocity (List[Any])

pub_twist_seq

• Purpose: Send a series of motion commands to the robot, enabling multi - step motion control.
• Parameters:
  • linear: List of linear velocities (List[Any])
  • angular: List of angular velocities (List[Any])
  • duration: Time interval for each step (in seconds) (List[Any])

🚀 Quick Start

1. Set the IP and port to connect to rosbridge.

• Open server.py and change your LOCAL_IP, ROSBRIDGE_IP, and ROSBRIDGE_PORT. (The default value of ROSBRIDGE_PORT is 9090)

2. Run the rosbridge server.

For ROS 1:

roslaunch rosbridge_server rosbridge_websocket.launch

For ROS 2:

ros2 launch rosbridge_server rosbridge_websocket_launch.xml

3. Run any AI system with ros-mcp-server imported.

4. Input "Let the robot move forward."

5. Check the status of rosbridge_server and ROS.

Command:

roslaunch rosbridge_server rosbridge_websocket.launch

Example output:

... Done ...

6. The test is completed. The robot will perform corresponding actions based on the received natural - language instructions.