Skip to content

README.md

Latest commit

 

History

History
58 lines (43 loc) · 1.6 KB

README.md

File metadata and controls

58 lines (43 loc) · 1.6 KB
Raw

ROS2 MCP Agent

A ROS2 agent that uses Model Context Protocol (MCP) for modular tool management.

Architecture

┌────────────────┐      ┌────────────────┐      ┌────────────────┐
│   MCP Agent    │─────▶│   MCP Server   │─────▶│     ROS2       │
│  (LLM Client)  │◀─────│  (ROS2 Tools)  │◀─────│    System      │
└────────────────┘      └────────────────┘      └────────────────┘

Features

  • MCP Protocol: Standard interface for LLM-tool communication
  • ROS2 Introspection: List topics, nodes, services
  • Topic I/O: Read and publish to any topic
  • Service Calls: Call any ROS2 service
  • Modular: Add more MCP servers for new capabilities

Quick Start

# Install dependencies
pip install mcp langchain langchain-together

# Set API key
export TOGETHER_API_KEY="your-key"

# Build
cd ~/rs_ws
colcon build --packages-select ros_vlm
source install/setup.bash

# Run demo
ros2 launch ros_vlm mcp_demo.launch.py

# Send command
ros2 topic pub /agent_input std_msgs/msg/String "data: 'list topics'" --once

Project Structure

ros_vlm/
├── mcp_agent_node.py   # MCP client agent
├── mcp_server.py       # ROS2 tools server
└── tools/
    └── ros2_tools.py   # Tool implementations

LLM Providers

Set LLM_PROVIDER env var:

  • together (default)
  • groq
  • openai