13. README.md

Robocore

Robocore is a pure Python middleware for robotic communication, emphasizing simplicity, customizability, and ease of use. It enables high-bandwidth, low-latency communication suitable for streaming data like RGB-D arrays.

Philosophy

Robocore aims to lower the barrier to entry for robotic communication by providing an easy-to-use and highly customizable framework. The goal is to make sure that you can focus on your methods without worrying about the underlying communication complexities.

Features

• Publish/Subscribe Paradigm: Simple string-based topics with support for multiple publishers and subscribers.
• Centralized Key-Value Store: A database-like system for storing and retrieving key-value pairs.
• Automatic Server Initiation: Servers for topic registry and key-value store start automatically in the background.
• Visualization Tools: Real-time data inspection and visualization using rerun-sdk.
• Configuration Files: Use .yaml files to specify topic types and additional info.

Installation

pip install robocore

Getting Started

Simple Publish/Subscribe

import robocore as rc

def publisher():
    pub = rc.Publisher(topic="camera_feed")
    pub.publish(data)  # Replace 'data' with your data
    pub.close()

def subscriber():
    sub = rc.Subscriber(topic="camera_feed")
    data = sub.receive()
    sub.close()

Multiple Subscribers

Check out the example in the examples folder:

python examples/multiple_subscribers.py

Robotic Manipulator Example

Simulate a robotic manipulator with multiple publishers and subscribers:

python examples/robotic_manipulator.py

Cross-Device Communication

Use Tailscale for networking across devices:

• Install Tailscale on both devices.
• Run the publisher on one device and the subscriber on the other.
• Modify the connect method to use the Tailscale IP address.

Configuration

Use a .yaml file to specify topic types and visualization settings:

topics:
  - name: camera_feed
    type: image
    visualize: true

Visualization

Integrate with rerun-sdk for real-time visualization:

visualizer = rc.Visualizer(topics=["camera_feed"])
visualizer.start()

Contributing

Contributions are welcome! Please refer to the documentation and follow the coding standards.

For more examples and detailed usage, please refer to the examples folder.