This repository serves as a comprehensive, curated collection of resources at the intersection of Large Language Models (LLMs) and Robotics. It targets researchers, engineers, and students interested in leveraging LLMs for robot control, planning, perception, and simulation, offering a structured overview of foundational concepts, cutting-edge research papers, and practical project demonstrations.

How It Works

The repository organizes a vast array of LLM-robotics research by functional area, including reasoning, planning, manipulation, instruction following, navigation, simulation frameworks, and perception. It links to seminal papers, code repositories, and educational materials, providing a structured learning path and a reference for current advancements. The curation emphasizes practical applications and theoretical underpinnings, facilitating the exploration of how LLMs can imbue robots with enhanced understanding and control capabilities.

Quick Start & Requirements

This repository is a curated list of links and does not require installation or execution. It serves as a knowledge base.

Highlighted Details

• Extensive coverage of research papers from major conferences (CoRL, CVPR, ICML, ICLR, IROS, NeurIPS) and arXiv.
• Links to code implementations for many research projects, facilitating reproducibility and further development.
• Inclusion of educational resources, from foundational transformer courses to specialized LLM-robotics classes.
• Demonstrations of real-world LLM-powered robots and hackathon projects, showcasing practical applications.

Maintenance & Community

The repository is maintained by jrin771. Community engagement and updates are primarily driven through GitHub.

Licensing & Compatibility

The repository itself is licensed under an unspecified license on GitHub. Individual linked research papers and code repositories retain their original licenses. Compatibility for commercial use depends on the licenses of the linked projects.

Limitations & Caveats

As a curated list, the repository does not provide executable code or a unified framework. The rapid pace of LLM and robotics research means content may require continuous updates to remain fully current. Some linked code repositories may be outdated or have limited maintenance.