Skip to content

jbaumgartl/go2-webrtc

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

47 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

go2-webrtc - WebRTC API or Unitree GO2 Robots

The go2-webrtc project provides a WebRTC API for Unitree GO2 Robots, enabling real-time communication and control over these robots through a web interface. This project simplifies the process of connecting to and controlling Unitree GO2 Robots by leveraging the WebRTC protocol for efficient, low-latency communication.

Go2's WebRTC API supports all models, including Go2 Air, Pro, and Edu versions.

Features

  • WebRTC Integration: Utilizes WebRTC to establish a real-time communication channel between the web client and the robot.
  • User Interface: Includes a simple web interface for connecting to the robot, sending commands, and viewing the robot's video stream.
  • Command Execution: Allows users to execute predefined commands on the robot, such as movement and action commands.
  • Persistent Settings: Saves connection settings (token and robot IP) in the browser's localStorage for easier reconnection.

Getting Started

To get started with go2-webrtc, clone this repository and serve the index.html file from the backend server (server.py) to a modern web browser. Ensure that your Unitree GO2 Robot is powered on and connected to the same network as your computer.

git clone https://github.com/tfoldi/go2-webrtc
cd go2-webrtc
pip install -r python/requirements.txt
cd javascript
python ./server.py

Sample Frontend Application

The javascript folder contains a sample frontend application that demonstrates how to use the JS WebRTC API to connect to and control the Unitree GO2 Robots. Explore the javascript folder for more details and examples.

Python API for Backend/Console Applications

For backend or console applications, the python folder provides a Python API that interfaces with the Unitree GO2 Robots. This API is ideal for developers looking to integrate robot control into their Python applications or scripts. Check out the python folder for documentation and examples.

Prerequisites

  • A Unitree GO2 Robot accessible over the local network. All models supported including Air, Pro and Edu
  • Local network connection (STA-L) to the robot

Obtaining security token

Connecting to your device without a security token is possible and might allow a connection to be established. However, this method limits you to a single active connection at any time. To simultaneously use multiple clients, such as a WebRTC-based application and a phone app, a valid security token is necessary. This ensures secure, multi-client access to your device.

One way is to sniff the traffic between the dog and your phone. Assuming that you have Linux or Mac:

  1. Run tinyproxy or any other HTTP proxy on your computer
  2. Set your computer's IP and port as HTTP proxy on your phone
  3. Run wireshark or ngrep on your box sniffing port 8081 like ngrep port 8081.
  4. Look for the token in the TCP stream after you connect your phone to the dog via the app

The token looks like this in the request payload:

{
    "token": "eyJ0eXAiOizI1NiJtlbiI[..]CI6MTcwODAxMzUwOX0.hiWOd9tNCIPzOOLNA",
    "sdp": "v=0\r\no=- ",
    "id": "STA_localNetwork",
    "type": "offer"
}

Another option is to obtain token via the /login/email endpoint.

curl -vX POST https://global-robot-api.unitree.com/login/email -d "email=<EMAIL>&password=<MD5 hash of your password>"

Development

This project is structured around several key JavaScript files for the frontend and a Python package for backend or console applications. To contribute or modify the project, refer to these resources for implementing additional features or improving the existing codebase. PRs are welcome.

License

This project is licensed under the BSD 2-clause License - see the LICENSE file for details.

About

WebRTC API for Unitree GO2 Robots

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 50.4%
  • JavaScript 44.4%
  • HTML 5.2%