OpenRacer

OpenRacer is an open-source platform that empowers you to train machine-learning models for autonomous racing. Built with Python and powered by a Unity-based 3D environment, OpenRacer offers a dynamic and accessible way to develop and test AI models in a simulated racing environment.

🚀 Features

• Python-Powered Training: Use Python to develop, train, and fine-tune your machine learning models.
• 3D Unity Environment: Experience immersive simulations with our Unity-based 3D racing environment, which connects to Python via WebSocket.
• Custom Track Creation: Generate your own tracks using coordinates provided by the server, or use premade tracks.
• Real-Time Progress Monitoring: Visualize your model's training progress and racing performance through an integrated web interface with live updates and detailed graphs.
• Hyperparameter Tuning: Easily adjust hyperparameters to optimize your model's performance.
• Open-Source Flexibility: Customize and extend the platform to suit your unique needs, thanks to its open-source codebase.

🎯 Impact

"Democratizing Autonomous Racing: Empowering Innovators and Learners to Shape the Future of AI."

OpenRacer is designed to lower the barrier to entry for autonomous racing and machine learning. Whether you're a hobbyist, a student, or a professional, our platform provides the tools you need to experiment, learn, and innovate in the field of AI.

📚 Getting Started

A stable version for wider use will be released soon

Nightly Installation

Prerequisites :

• Python 3.7+
• Unity 2022.3
• npm 9.5.0

Steps :

1. Clone the repository:

   git clone https://github.com/Loony4Logic/OpenRacer
   cd openracer

2. Install Python dependencies:

   cd "Python server"
   pip install -r requirements.txt

3. Run Python server

   This will run an exmaple model. Which is currently random. so don't expect it to give any good performace

   python server.py 

4. Open the Unity project:

   • Open Unity Hub, click on "Add Project," and select the OpenRacer folder within the project directory.
   • Load the scene you want to try. Training for training model. Testing for evalulating your model.

5. Start the web interface:

   cd React UI
   npm install
   npm run dev

🤝 Contributing

Contributions are welcome! Please create an issue or contact me through LinkedIn. I will add structured issues and guidelines after a while.

🛡️ License

This project is licensed under the MIT License - see the LICENSE file for details.

🙌 Acknowledgments

• Inspired by AWS DeepRacer.
• Special thanks to Tejas Nair for 3D model and Palak Agarwal.

---