End-to-End Autonomous Vehicle Driving Based on Text-Based Instructions

GitHub | Website | Video | Meeting Logs | Model Card | Data Card

About the project

🚀 Project Goal: Integrating Large Language Models with end-to-end autonomous driving models. This project is supported by 2024 Google Summer of Code and JdeRobot.

Project Structure

Each folder contains related documentation.

• carla_llms This folder includes an end-to-end demo generated based on large language models and validated in CARLA.

• data_parsing This directory contains the original development version, used for standalone testing of all scripts. It serves as the core environment for debugging and optimizing the functionality of the scripts.

• streamlit This is the web-based streamlit app, packaged for online deployment using the Streamlit framework.

• carla This directory integrates the model with the CARLA simulation for testing and validation.

• docs Documentation files related to the project, including blogs and technical references.

Acknowledgments

• gsoc2023-Meiqi_Zhao